INDEX

Explanations

coding commands and structure related to data storage and manipulation

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

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

 Majefty

-1.09

 itſelf

-1.09

 Jefus

-1.05

Efq

-1.04

ſelf

-1.02

 pleaſure

-0.97

 ―――――

-0.96

 iſt

-0.96

 doubtnut

-0.94

ſelves

-0.94

POSITIVE LOGITS

0.59

0.52

Fake

0.52

0.51

0.50

_{

0.47

 gestes

0.47

wsi

0.46

Activations Density 0.004%

coding commands and structure related to data storage and manipulation

No Comments

No Known Activations