INDEX

Explanations

phrases or clauses within parentheses

oai_token-act-pair · gpt-3.5-turbo

opening parentheses in various contexts

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

 comparisons

-0.75

 cockpit

-0.66

Cly

-0.65

 Prism

-0.65

 abroad

-0.65

 maker

-0.65

 Eston

-0.65

 tack

-0.64

 overhaul

-0.64

 footprint

-0.64

POSITIVE LOGITS

mostly

1.58

possibly

1.55

usually

1.48

sic

1.48

almost

1.42

formerly

1.36

partial

1.35

insert

1.35

probably

1.34

...)

1.30

Activations Density 0.075%

phrases or clauses within parentheses

opening parentheses in various contexts

No Comments

No Known Activations