INDEX

Explanations

technical terms related to electronics and communication technologies.

oai_token-act-pair · gpt-4o-mini Triggered by @bot

technical information or troubleshooting steps related to computer hardware or software.

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

words or phrases related to proximity or closeness, particularly the word "near" and related concepts like "near-field", "near-term", or "nearby".

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @bot

"near"

np_acts-logits-general · gemini-2.0-flash

near future or vicinity

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Mundi

-0.59

tungs

-0.59

riwal

-0.56

googleapis

-0.54

TAWA

-0.53

ğraf

-0.53

gués

-0.52

CWE

-0.52

schule

-0.50

 ddelweddau

-0.50

POSITIVE LOGITS

sighted

0.99

 resourceCulture

0.94

near

0.88

 NEAR

0.82

 near

0.80

NEAR

0.78

 Near

0.78

fromnode

0.73

 vicinity