INDEX

Explanations

conversational dialogue or interactions between individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

MPN

-0.63

writeValue

-0.59

 BrowserModule

-0.58

prnewswire

-0.57

KURZBESCHREIBUNG

-0.54

สือ

-0.53

 keyPressed

-0.52

 AssemblyCulture

-0.51

AnimationsModule

-0.50

Gambas

-0.49

POSITIVE LOGITS

 suggested

0.78

 suggesting

0.76

 told

0.74

 said

0.73

说是

0.73

 assured

0.71

 explained

0.71

 mentioning

0.70

 insisted

0.69

 mention

0.68

Activations Density 0.316%

conversational dialogue or interactions between individuals

No Comments

No Known Activations