INDEX

Explanations

terms related to evaluation and determination of legality or compliance

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

"[word] to be" or "[word] to"

np_acts-logits-general · gemini-2.0-flash

{ "explanation": "found to"}

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

AndEndTag

-0.54

ọa

-0.54

soort

-0.53

ofire

-0.51

MethodManager

-0.46

呼ば

-0.45

findpost

-0.45

 vergessen

-0.45

veloped

-0.45

 đốc

-0.43

POSITIVE LOGITS

 capable

0.86

 worthy

0.85

 essere

0.82

 unsuitable

0.82

 être

0.82

tobe

0.79

 unworthy

0.76

 incapable

0.76

 responsible

0.72

 suitable

0.71

Activations Density 0.496%

terms related to evaluation and determination of legality or compliance

"[word] to be" or "[word] to"

{ "explanation": "found to"}

No Comments

No Known Activations