INDEX

Explanations

mentions of reviews

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 ProtoMessage

-0.56

Sucesor

-0.52

 Wicidata

-0.51

NewUrlParser

-0.51

XDECREF

-0.49

ChromeDriver

-0.49

anius

-0.49

 Infórmanos

-0.49

posedge

-0.48

tahui

-0.48

POSITIVE LOGITS

 reviews

0.66

 review

0.60

 reviewer

0.60

 reviewers

0.59

 Reviews

0.58

Reviews

0.56

 ratings

0.56

 Review

0.55

 Ratings

0.55

Review

0.53

Activations Density 0.022%

mentions of reviews

No Comments

No Known Activations