INDEX

Explanations

references to different periods of time or eras

oai_token-act-pair · gpt-3.5-turbo

references to different historical periods or "times."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

CHAT

-0.97

ITAL

-0.81

ramid

-0.80

RD

-0.79

Reviewer

-0.76

ARI

-0.74

irm

-0.74

thereum

-0.73

Solid

-0.73

ItemTracker

-0.73

POSITIVE LOGITS

cale

1.28

manship

0.84

cape

0.80

hare

0.78

mith

0.75

pan

0.75

eries

0.74

hower

0.73

cens

0.73

 elapsed

0.71

Activations Density 0.024%

references to different periods of time or eras

references to different historical periods or "times."

No Comments

No Known Activations