INDEX

Explanations

profanity and aggressive language

oai_token-act-pair · gpt-3.5-turbo

expressions of strong negative emotions or insults

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

ItemImage

-0.80

Footnote

-0.77

PDF

-0.73

 cautiously

-0.71

 sharply

-0.70

Alternatively

-0.69

 concess

-0.69

 emphas

-0.69

*)

-0.68

 outset

-0.67

POSITIVE LOGITS

 fuckin

1.90

 shit

1.60

 fucking

1.56

 fuck

1.54

fuck

1.52

 fucked

1.48

 FUCK

1.47

 Fuck

1.42

Fuck

1.40

nig

1.39

Activations Density 1.052%

profanity and aggressive language

expressions of strong negative emotions or insults

No Comments

No Known Activations