INDEX

Explanations

punctuation marks, specifically periods

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

RegressionTest

-0.62

 betweenstory

-0.57

发表于

-0.55

 ModelExpression

-0.54

 Wikiseite

-0.53

новниш

-0.52

HXLINE

-0.50

ManagedBean

-0.48

enterOuterAlt

-0.47

MessageTagHelper

-0.47

POSITIVE LOGITS

 collections

0.68

 collection

0.61

 Collections

0.56

 computing

0.52

collections

0.52

 Collection

0.51

Collections

0.50

 lists

0.49

Collection

0.47

Lis

0.47

Activations Density 0.000%

punctuation marks, specifically periods

No Comments

No Known Activations