INDEX

Explanations

interrogative sentences with question marks

oai_token-act-pair · gpt-3.5-turbo

rhetorical questions and queries regarding opinions or expectations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 enriched

-0.84

 synerg

-0.81

 downstream

-0.80

 flowing

-0.79

 sustained

-0.79

 netted

-0.79

 raided

-0.79

 flowering

-0.79

 neighb

-0.78

 upstream

-0.77

POSITIVE LOGITS

Answer

2.42

Well

2.02

Absolutely

1.82

Yes

1.75

Probably

1.72

Wr

1.64

Actually

1.56

Certainly

1.55

YES

1.54

Maybe

1.51

Activations Density 0.152%

interrogative sentences with question marks

rhetorical questions and queries regarding opinions or expectations

No Comments

No Known Activations