INDEX

Explanations

references to statistical or empirical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/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.12.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

GEBURTSDATUM

-0.89

 ―――――

-0.87

 dieß

-0.85

 himo

-0.85

 öffnungszeiten

-0.84

 noDo

-0.80

Tikang

-0.79

 pleaſure

-0.78

 auffi

-0.75

 niedersachsen

-0.75

POSITIVE LOGITS

1.47

mathrm

0.70

0.69

|}{$

0.61

setcounter

0.61

0.58

uarkan

0.57

enterOuterAlt

0.57

フェクト

0.56

Activations Density 1.651%

references to statistical or empirical data

No Comments

No Known Activations