INDEX

Explanations

words or phrases indicating a comparison or contrast between different scenarios or possibilities

oai_token-act-pair · gpt-3.5-turbo

phrases indicating uncertainty or qualification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Reviewer

-0.66

FTWARE

-0.66

itivity

-0.64

HTTP

-0.58

DonaldTrump

-0.56

visual

-0.56

View

-0.55

Owner

-0.55

Package

-0.55

SIGN

-0.55

POSITIVE LOGITS

 least

1.62

mosp

1.14

onement

1.06

 times

0.96

yp

0.90

roph

0.90

ention

0.87

hens

0.85

 odds

0.85

letico

0.85

Activations Density 0.094%

words or phrases indicating a comparison or contrast between different scenarios or possibilities

phrases indicating uncertainty or qualification

No Comments

No Known Activations