INDEX

Explanations

references to temporary actions or conditions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "temporary."

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

ioch

-0.79

sonian

-0.79

VID

-0.78

bold

-0.78

riad

-0.77

eus

-0.76

gans

-0.75

etus

-0.75

ICLE

-0.74

Jenn

-0.73

POSITIVE LOGITS

 restraining

1.16

 accommodation

1.00

 temporary

0.91

 lodging

0.90

porary

0.89

 inval

0.87

 truce

0.87

 halting

0.83

 setback

0.82

 temporarily

0.81

Activations Density 0.015%

references to temporary actions or conditions

instances of the word "temporary."

No Comments

No Known Activations