INDEX

Explanations

phone-related words or phrases

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "phone."

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

rix

-1.13

urities

-0.73

iverpool

-0.71

liga

-0.70

 Inher

-0.70

nesday

-0.69

Wem

-0.65

alis

-0.65

xual

-0.64

riks

-0.64

POSITIVE LOGITS

 phones

0.88

 phone

0.85

ysis

0.84

 contacts

0.84

phones

0.83

gap

0.81

 booths

0.79

 booth

0.78

 handset

0.77

 calls

0.74

Activations Density 0.022%

phone-related words or phrases

occurrences of the word "phone."

No Comments

No Known Activations