INDEX

Explanations

references to shock

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 raiſ

-0.69

 ſeveral

-0.69

 myſelf

-0.69

 whoſe

-0.68

 themſelves

-0.68

 fubject

-0.64

SError

-0.63

Personendaten

-0.63

rungsseite

-0.63

uſ

-0.62

POSITIVE LOGITS

 shock

0.73

InjectAttribute

0.68

 kann

0.66

 wird

0.65

 shocks

0.63

shock

0.61

 SHOCK

0.58

 konnte

0.58

 Dong

0.58

Shock

0.56

Activations Density 0.069%

references to shock

No Comments

No Known Activations