INDEX

Explanations

terms related to strikes or attacks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

робнее

-0.77

engkapnya

-0.72

phorbia

-0.71

HomeAsUpEnabled

-0.71

escens

-0.69

 carpeta

-0.67

 Glen

-0.66

 Clarence

-0.65

Entities

-0.65

 Mess

-0.64

POSITIVE LOGITS

 strike

1.85

 Strikes

1.81

 strikes

1.80

 Strike

1.76

Strike

1.75

strike

1.67

 striking

1.49

 struck

1.36

 Striking

1.27

stri

1.20

Activations Density 0.105%

terms related to strikes or attacks

No Comments

No Known Activations