INDEX

Explanations

negative observations about visibility and experiences related to environmental conditions

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

blurry photos

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

wapV

-0.56

acceptance

-0.53

DoubleQuotes

-0.53

 avoient

-0.49

Acceptance

-0.48

PreferredItem

-0.46

approval

-0.46

oil

-0.46

Hentet

-0.46

TokenNameLPAREN

-0.45

POSITIVE LOGITS

setVerticalGroup

0.54

ngdoc

0.41

 disappointing

0.35

 jadx

0.35

 dull

0.34

enderror

0.33

 boring

0.32

MTL

0.32

 ciudadana

0.31

 Kraj

0.31

Activations Density 0.084%

negative observations about visibility and experiences related to environmental conditions

blurry photos

No Comments

No Known Activations