INDEX

Explanations

proper nouns, especially focusing on the surname "Johnston"

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals, particularly those named Johnston or Russo

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ö¼

-0.78

ŀ

-0.73

awaru

-0.71

oiler

-0.70

gemony

-0.66

bidden

-0.64

BOOK

-0.64

ãĤ£

-0.63

Ü

-0.63

 princ

-0.63

POSITIVE LOGITS

 Johnston

1.49

ston

1.17

stown

0.98

shire

0.87

patrick

0.83

ansson

0.82

acre

0.82

igan

0.76

 Kingston

0.73

 Vaughan

0.72

Activations Density 0.008%

proper nouns, especially focusing on the surname "Johnston"

mentions of specific individuals, particularly those named Johnston or Russo

No Comments

No Known Activations