INDEX

Explanations

articles or news related to specific events or major occurrences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating significant events or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

©¶æ

-0.82

©¶æ¥µ

-0.80

ĸļ

-0.78

ccording

-0.75

 tremend

-0.72

 Haku

-0.71

Hok

-0.69

ŃĶ

-0.61

é¾

-0.61

 Caption

-0.60

POSITIVE LOGITS

cum

0.68

ennis

0.66

ove

0.62

psc

0.62

fuck

0.61

under

0.61

arters

0.60

anqu

0.60

artments

0.59

avez

0.59

Activations Density 0.095%

articles or news related to specific events or major occurrences

phrases indicating significant events or recommendations

No Comments

No Known Activations