INDEX

Explanations

phrases related to physical locations or people's names

oai_token-act-pair · gpt-3.5-turbo

words related to a state of being known as "being scared" or "fear."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Mit

-0.63

OD

-0.62

âĶĢâĶĢ

-0.59

 rooting

-0.57

 Crew

-0.57

 Bulldogs

-0.56

YS

-0.56

 Pavilion

-0.56

INS

-0.56

annels

-0.56

POSITIVE LOGITS

paren

1.02

thro

0.97

bryce

0.85

throp

0.81

atche

0.77

baugh

0.77

ndra

0.76

eer

0.75

tsky

0.74

ner

0.72

Activations Density 0.009%

phrases related to physical locations or people's names

words related to a state of being known as "being scared" or "fear."

No Comments

No Known Activations