INDEX

Explanations

the word "and" in various contexts

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

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

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mergeFrom

-0.52

 estimés

-0.46

 implication

-0.44

 tarko

-0.44

</strong>

-0.43

opengl

-0.43

GeneratedCode

-0.42

ransition

-0.42

 cual

-0.41

Fre

-0.41

POSITIVE LOGITS

 للمعارف

0.69

ArgsConstructor

0.65

tvguidetime

0.60

Personendaten

0.60

AttributeSet

0.59

 instr

0.58

 typelib

0.58

SharedDtor

0.58

 propOrder

0.57

ivez

0.57

Activations Density 0.002%

the word "and" in various contexts

No Comments

No Known Activations