INDEX

Explanations

elements related to intermediate steps or components in processes or explanations

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

localctx

-0.36

GenerationType

-0.36

gonic

-0.36

ffions

-0.34

 ordinaires

-0.33

RuleContext

-0.31

seitigen

-0.31

yyj

-0.30

DataAnnotations

-0.30

JpaRepository

-0.30

POSITIVE LOGITS

 intermediate

0.79

intermediate

0.77

Intermediate

0.69

 Intermediate

0.68

 intermediates

0.63

 intermédiaire

0.57

 intermediary

0.57

 intermedi

0.56

 intermediaries

0.54

 Zwischen

0.51

Activations Density 1.169%

elements related to intermediate steps or components in processes or explanations

No Comments

No Known Activations