INDEX

Explanations

mention

np_max-act-logits · gemini-2.0-flash

words related to quantities, amounts, and metrics.

oai_token-act-pair · deepseek-v3 Triggered by @finnwit

Verbs and nouns describing actions related to communication, analysis, or descriptive content.

eleuther_acts_top20 · deepseek-v3 Triggered by @finnwit

contains (first token in MAX_ACTIVATING_TOKENS)

np_max-act · deepseek-v3 Triggered by @finnwit

mention

np_max-act-logits · deepseek-v3 Triggered by @finnwit

Explanation could not be parsed.

eleuther_acts_top20 · deepseek-r1 Triggered by @finnwit

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

weight

-0.28

odia

-0.28

phase

-0.26

à¸Ķà¸²à¸§

-0.25

åıĺå½¢

-0.24

 whistle

-0.24

Stroke

-0.24

 weight

-0.23

å°ī

-0.23

åŀĭ

-0.23

POSITIVE LOGITS

 mention

0.88

è°Īè®º

0.82

 mentions

0.82

 mentioning

0.81

æıĲåĪ°

0.81

æıĲåıĬ

0.80

è°ĪåıĬ

0.76

 Mention

0.72

è°ĪåĪ°

0.71

 discussing

0.68

Activations Density 0.001%

mention

words related to quantities, amounts, and metrics.

Verbs and nouns describing actions related to communication, analysis, or descriptive content.

contains (first token in MAX_ACTIVATING_TOKENS)

mention

Explanation could not be parsed.

No Comments

No Known Activations

mention

words related to quantities, amounts, and metrics.

Verbs and nouns describing actions related to communication, analysis, or descriptive content.

contains (first token in MAX_ACTIVATING_TOKENS)

mention

Explanation could not be parsed.

No Comments

No Known Activations