INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

entities identified as individuals or professionals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tumblr

-0.95

DragonMagazine

-0.75

 Marketable

-0.68

 derailed

-0.64

 manifesto

-0.63

 totality

-0.63

 proved

-0.63

 lasts

-0.62

 constitutes

-0.61

pmwiki

-0.61

POSITIVE LOGITS

emon

0.83

anton

0.78

 Choi

0.77

 Angelo

0.74

 Rank

0.72

Lee

0.71

uria

0.71

 Cohen

0.71

 Shapiro

0.70

 Hawkins

0.70

Activations Density 0.178%

names of individuals

entities identified as individuals or professionals

No Comments

No Known Activations