INDEX

Explanations

mentions of individuals involved in narratives or trials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

itself

-0.61

 itſelf

-0.58

 itself

-0.54

LookAnd

-0.54

 contenir

-0.52

 Itself

-0.51

MessageTagHelper

-0.50

rovnik

-0.49

它

-0.48

 فريبيس

-0.48

POSITIVE LOGITS

 whom

1.20

whom

0.97

who

0.87

がいる

0.78

 whose

0.78

Whom

0.73

 Whom

0.68

whose

0.66

who

0.63

 names

0.61

Activations Density 0.707%

mentions of individuals involved in narratives or trials

No Comments

No Known Activations