INDEX

Explanations

references to contextual elements or qualifications in discussions

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

Followed by a comma and then "with"

np_acts-logits-general · gemini-2.0-flash

appended descriptive clauses

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

ってお

-0.41

мети

-0.39

等

-0.38

=%

-0.37

ths

-0.37

 parle

-0.37

代わりに

-0.37

-0.36

jesz

-0.36

POSITIVE LOGITS

 which

1.29

 designed

1.06

 whose

1.05

 capable

1.02

which

1.00

 consisting

0.98

whose

0.95

 aimed

0.90

 featuring

0.88

 destined

0.88

Activations Density 0.515%

references to contextual elements or qualifications in discussions

Followed by a comma and then "with"

appended descriptive clauses

No Comments

No Known Activations