INDEX

Explanations

phrases indicating belief, general acceptance, or attribution of ideas to specific individuals or experts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

ſures

-0.45

 hany

-0.42

Té

-0.41

 intran

-0.40

addContainerGap

-0.40

 Rans

-0.40

Par

-0.40

Par

-0.40

 Hello

-0.39

Doll

-0.39

POSITIVE LOGITS

 believed

0.72

 вважа

0.63

 считается

0.62

principalColumn

0.61

 considered

0.59

 dianggap

0.58

 believe

0.57

 považ

0.57

 Considered

0.55

 считают

0.55

Activations Density 0.591%

phrases indicating belief, general acceptance, or attribution of ideas to specific individuals or experts

No Comments

No Known Activations