INDEX

Explanations

phrases indicating preference or substitution

oai_token-act-pair · gpt-3.5-turbo

phrases indicating preference or choices in favor of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

estate

-0.75

iana

-0.69

caster

-0.68

"]=>

-0.67

Initialized

-0.67

>[

-0.66

affected

-0.65

problem

-0.63

iott

-0.63

 proble

-0.63

POSITIVE LOGITS

livion

0.79

 manageable

0.71

 rubble

0.69

azel

0.67

 nearest

0.64

 piles

0.63

 pasture

0.62

 plain

0.61

 paste

0.60

Tier

0.59

Activations Density 0.566%

phrases indicating preference or substitution

phrases indicating preference or choices in favor of something

No Comments

No Known Activations