INDEX

Explanations

dates in the format "Month.Day"

oai_token-act-pair · gpt-3.5-turbo

punctuation, specifically the period at the end of sentences

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

atories

-0.66

OGR

-0.66

 illiter

-0.60

 outstanding

-0.59

 intellectual

-0.58

 achievements

-0.58

 clut

-0.58

 chimpanzees

-0.57

 achievement

-0.57

 intangible

-0.56

POSITIVE LOGITS

 Madness

0.87

Downloadha

0.87

furt

0.83

0.80

0.75

 Fool

0.75

0.73

ember

0.73

Activations Density 0.029%

dates in the format "Month.Day"

punctuation, specifically the period at the end of sentences

No Comments

No Known Activations