INDEX

Explanations

elements related to competition and evaluation criteria

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

winning or victory

np_acts-logits-general · gemini-2.0-flash

win or victory

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

исленность

-0.60

czaj

-0.58

ArgumentParser

-0.49

IMPORTED

-0.46

ibatis

-0.45

NOSTIC

-0.45

Superview

-0.45

ptonshire

-0.45

itest

-0.44

 useCallback

-0.43

POSITIVE LOGITS

win

3.30

 winning

3.19

 wins

3.09

 victory

2.93

 winner

2.83

 winners

2.73

win

2.70

Winning

2.70

winning

2.67

 Winning

2.65

Activations Density 0.539%

elements related to competition and evaluation criteria

winning or victory

win or victory

No Comments

No Known Activations