INDEX

Explanations

phrases related to reactions and responses to actions or events

oai_token-act-pair · gpt-3.5-turbo

phrases indicating responses to events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Inher

-0.69

 Forsaken

-0.63

 "$:/

-0.63

 ceilings

-0.62

inav

-0.62

 Archdemon

-0.62

 ledger

-0.61

teenth

-0.60

 visibility

-0.59

ritten

-0.59

POSITIVE LOGITS

onse

0.82

0.80

orus

0.70

universal

0.69

0.66

 prov

0.65

oxide

0.65

xit

0.64

atre

0.64

 reply

0.64

Activations Density 0.169%

phrases related to reactions and responses to actions or events

phrases indicating responses to events or actions

No Comments

No Known Activations