INDEX

Explanations

instances of numerical references or identifiers

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

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

rimir

-0.49

LAHOMA

-0.49

 odkazy

-0.49

atecas

-0.48

snackbar

-0.46

ברס

-0.46

diyesi

-0.45

 Snackbar

-0.44

acabana

-0.44

 sensibili

-0.43

POSITIVE LOGITS

ftype

3.59

 TType

0.99

qtype

0.88

ntype

0.82

 wireType

0.82

 type

0.80

FunctionType

0.78

 types

0.75

 TIPO

0.74

type

0.73

Activations Density 0.000%

instances of numerical references or identifiers

No Comments

No Known Activations