INDEX

Explanations

keywords related to reports, articles, and presentations

oai_token-act-pair · gpt-3.5-turbo

references to formal reports and articles

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

tics

-0.85

 Cups

-0.84

ernels

-0.82

artifacts

-0.78

ses

-0.78

itudes

-0.76

eworks

-0.76

 trophies

-0.71

 Empires

-0.70

 races

-0.70

POSITIVE LOGITS

 titled

0.85

 called

0.78

onym

0.77

 consisting

0.74

 Called

0.73

atical

0.72

 headlined

0.71

 dubbed

0.69

 resembling

0.69

 reminiscent

0.69

Activations Density 0.377%

keywords related to reports, articles, and presentations

references to formal reports and articles

No Comments

No Known Activations