INDEX

Explanations

medical and investigative terms related to examinations and determinations

oai_token-act-pair · gpt-3.5-turbo

references to investigations or inquiries related to criminal activity

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

pload

-0.61

 routers

-0.58

illas

-0.58

 noodles

-0.56

 centrif

-0.55

usercontent

-0.55

 islands

-0.54

 malls

-0.53

sod

-0.53

 jets

-0.52

POSITIVE LOGITS

 unbiased

0.69

 practition

0.66

isman

0.61

 Award

0.60

zsche

0.60

 Journalism

0.60

£ı

0.59

ascript

0.58

ournal

0.58

¿½

0.57

Activations Density 1.428%

medical and investigative terms related to examinations and determinations

references to investigations or inquiries related to criminal activity

No Comments

No Known Activations