INDEX

Explanations

mentions of specific references or tagged entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.hook_mlp_out

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

 autorytatywna

-0.75

 Normdatei

-0.65

 Monfieur

-0.65

AndEndTag

-0.63

 pinulongan

-0.63

 صوتيه

-0.62

 itſelf

-0.61

 themſelves

-0.60

يكب

-0.59

enumi

-0.59

POSITIVE LOGITS

gonic

0.52

 EventArgs

0.51

 getItemId

0.50

Charsets

0.50

ubblica

0.48

 Throwable

0.48

RIV

0.48

nextLine

0.47

UrlResolution

0.47

 sonor

0.46

Activations Density 0.045%

mentions of specific references or tagged entities

No Comments

No Known Activations