INDEX

Explanations

terms related to processes or elements of scientific and technical explanation

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

sn

-0.51

 ogóle

-0.47

 exchange

-0.45

sn

-0.45

 capacidade

-0.44

 aikana

-0.44

معلومات

-0.43

 aspetto

-0.43

情况下

-0.43

 difp

-0.43

POSITIVE LOGITS

ThroughAttribute

0.84

 saites

0.81

 AssemblyProduct

0.78

 eventdata

0.75

 мәкал

0.73

PhysRevLett

0.71

PMailer

0.70

aarrggbb

0.70

expandindo

0.68

oredCriteria

0.67

Activations Density 0.295%

terms related to processes or elements of scientific and technical explanation

No Comments

No Known Activations