INDEX

Explanations

mentions of the word "note" or its variations

oai_token-act-pair · gpt-3.5-turbo

references to notable events or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

£ı

-0.85

Ļ½

-0.73

ruary

-0.71

Kate

-0.71

dimension

-0.69

ailability

-0.61

hips

-0.60

 disapp

-0.60

antha

-0.59

 dred

-0.59

POSITIVE LOGITS

OPLE

1.12

tsky

1.02

lete

0.99

ote

0.98

otes

0.95

chnology

0.95

chn

0.90

atro

0.82

opsis

0.80

ito

0.79

Activations Density 0.011%

mentions of the word "note" or its variations

references to notable events or items

No Comments

No Known Activations