INDEX

Explanations

occurrences of the ampersand symbol

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 itſelf

-1.09

 snippetHide

-0.96

^(@)

-0.90

Efq

-0.90

 myſelf

-0.86

 Jefus

-0.85

 himſelf

-0.84

 themſelves

-0.82

 Theſe

-0.82

 Houſe

-0.80

POSITIVE LOGITS

0.80

([

0.66

0.60

([]

0.59

(&

0.57

(_

0.55

((

0.53

็จ

0.53

(&(

0.51

Activations Density 0.029%

occurrences of the ampersand symbol

No Comments

No Known Activations