INDEX

Explanations

the word "no" followed by either a specific concept or a more general term

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the absence or non-existence 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

endar

-0.69

iership

-0.65

RAFT

-0.63

 Vinyl

-0.63

 Literary

-0.62

kinson

-0.62

ktop

-0.60

rious

-0.60

lass

-0.59

 Tactics

-0.59

POSITIVE LOGITS

xious

1.26

oses

0.91

ct

0.88

 longer

0.87

obs

0.85

 meaningful

0.79

 matter

0.78

ose

0.78

isy

0.76

oooooooo

0.74

Activations Density 0.100%

the word "no" followed by either a specific concept or a more general term

phrases indicating the absence or non-existence of something

No Comments

No Known Activations