INDEX

Explanations

statements with strong opinions or beliefs

oai_token-act-pair · gpt-3.5-turbo

instances of authority and response situations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Reynolds

-0.65

Vu

-0.63

 Block

-0.62

 valve

-0.60

ãĤ¬

-0.60

ilon

-0.59

 indexes

-0.58

 Bell

-0.57

 Bever

-0.57

olin

-0.57

POSITIVE LOGITS

 suddenly

0.90

uddenly

0.80

?!"

0.78

 disrespect

0.76

 blatantly

0.74

someone

0.72

ocious

0.72

?!

0.70

ammed

0.70

ebook

0.70

Activations Density 1.098%

statements with strong opinions or beliefs

instances of authority and response situations

No Comments

No Known Activations