INDEX

Explanations

terms related to cellular processes and components

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 drastic

-1.65

 recent

-1.61

 severe

-1.59

 cref

-1.54

 bipartisan

-1.53

 royal

-1.51

ige

-1.50

lementary

-1.50

cester

-1.49

 certain

-1.46

POSITIVE LOGITS

ĵ

3.92

¤

3.84

°

3.73

¢

3.72

Ħ

3.59

Ł

3.56

£

3.49

©

3.47

Ģ

3.46

µ

3.46

Activations Density 0.025%

terms related to cellular processes and components

No Comments

No Known Activations