INDEX

Explanations

mentions of specific historical figures or events

oai_token-act-pair · gpt-3.5-turbo

references to a specific entity or group, likely focused on a cultural or entertainment context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Cind

-0.69

 DeVos

-0.65

JPM

-0.64

istors

-0.64

aminer

-0.62

aples

-0.62

 Beir

-0.62

 killer

-0.61

 LAPD

-0.60

ivably

-0.58

POSITIVE LOGITS

Åį

1.29

Â·Â·

1.05

su

0.97

Å«

0.97

Äĵ

0.91

¬

0.90

nen

0.90

nin

0.86

jin

0.86

ãĥ³ãĤ¸

0.85

Activations Density 0.006%

mentions of specific historical figures or events

references to a specific entity or group, likely focused on a cultural or entertainment context

No Comments

No Known Activations