INDEX

Explanations

references or mentions of the source of something

oai_token-act-pair · gpt-3.5-turbo

references to a source being discussed or implicated

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

aeper

-0.94

ellen

-0.81

awks

-0.76

paio

-0.72

wagen

-0.71

achine

-0.71

Ħ¢

-0.70

eer

-0.69

ead

-0.69

ipop

-0.67

POSITIVE LOGITS

Fed

0.99

source

0.92

 sources

0.91

 source

0.90

books

0.88

finding

0.87

 whence

0.83

book

0.80

ãĥ¼ãĥĨãĤ£

0.79

 Sources

0.79

Activations Density 0.029%

references or mentions of the source of something

references to a source being discussed or implicated

No Comments

No Known Activations