INDEX

Explanations

concepts related to countering or opposing actions or arguments

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<?

-0.49

txtName

-0.48

makedirs

-0.48

 Comprometido

-0.47

 detalladas

-0.44

niña

-0.42

Біографія

-0.42

 AssemblyCulture

-0.42

 itulah

-0.42

 betreft

-0.42

POSITIVE LOGITS

 counter

0.91

 counteract

0.78

 countered

0.75

 Counter

0.73

counter

0.73

 against

0.73

 COUNTER

0.72

 contre

0.67

 opponent

0.66

 antidote

0.66

Activations Density 0.035%

concepts related to countering or opposing actions or arguments

No Comments

No Known Activations