INDEX

Explanations

adjectives associated with various scenes or topics

oai_token-act-pair · gpt-3.5-turbo

references to various forms of unexpected or self-inflicted harm

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

 strengthened

-0.53

 âĶľ

-0.51

 Aden

-0.51

earchers

-0.47

ravel

-0.47

 Mehran

-0.46

braska

-0.46

 leasing

-0.46

imaru

-0.45

 Modified

-0.45

POSITIVE LOGITS

?).

0.71

?".

0.69

thood

0.61

)).

0.61

$.

0.60

;)

0.57

 crap

0.54

 shit

0.53

ãģ¾

0.53

ãĤĭ

0.53

Activations Density 1.128%

adjectives associated with various scenes or topics

references to various forms of unexpected or self-inflicted harm

No Comments

No Known Activations