INDEX

Explanations

responses related to the initiation of a discussion or a new idea

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 ModelExpression

-0.76

 مشارکت‌کنندگان

-0.61

RUnlock

-0.61

+:+

-0.57

=$?

-0.56

 protoimpl

-0.54

 comfor

-0.53

 CHAPITRE

-0.52

>';

-0.51

 <=",

-0.51

POSITIVE LOGITS

これも

0.52

anches

0.51

InitVars

0.50

Transfer

0.49

transfer

0.49

TagHelper

0.48

NewGuid

0.48

 retire

0.47

Према

0.47

retired

0.47

Activations Density 0.556%

responses related to the initiation of a discussion or a new idea

No Comments

No Known Activations