INDEX

Explanations

programming and code-related structures or keywords

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 defaultstate

-0.58

providedIn

-0.57

oredCriteria

-0.57

DotNetBar

-0.56

 betweenstory

-0.55

localctx

-0.46

complexContent

-0.46

 MainAxisSize

-0.46

-0.45

TargetApi

-0.44

POSITIVE LOGITS

 atom

0.60

Atom

0.56

 ATOM

0.56

 Atom

0.54

atom

0.52

 term

0.52

ATOM

0.48

 Term

0.48

Term

0.47

 BEAM

0.47

Activations Density 0.591%

programming and code-related structures or keywords

No Comments

No Known Activations