INDEX

Explanations

references to final events or conclusions

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific name or entity associated with notable events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

hare

-0.64

anwhile

-0.63

 orally

-0.62

ĸļ

-0.61

tan

-0.61

theless

-0.60

é¾įå¥ĳå£«

-0.60

ktop

-0.58

 meanings

-0.58

 hint

-0.57

POSITIVE LOGITS

isher

1.09

ament

1.04

boa

1.04

ners

1.01

ishers

1.01

anced

1.00

ancing

0.99

itures

0.96

ivers

0.95

ned

0.92

Activations Density 0.009%

references to final events or conclusions

mentions of a specific name or entity associated with notable events

No Comments

No Known Activations