INDEX

Explanations

studies or research-related phrases

oai_token-act-pair · gpt-3.5-turbo

references to research studies and their findings

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

 Strongh

-0.75

 Petty

-0.70

 Loyal

-0.70

fty

-0.68

 Redemption

-0.65

atra

-0.64

ibles

-0.63

 Newt

-0.60

 Sundays

-0.60

die

-0.60

POSITIVE LOGITS

 studies

1.14

 examining

1.08

 evaluating

1.06

 conducted

1.02

 studying

1.00

 confirming

0.98

 estimating

0.97

 correl

0.96

 documenting

0.96

 demonstrating

0.96

Activations Density 0.181%

studies or research-related phrases

references to research studies and their findings

No Comments

No Known Activations