INDEX

Explanations

proper nouns or titles related to individuals or characters

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "man"

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

CAST

-0.77

ritical

-0.76

irtual

-0.74

rss

-0.73

¥ŀ

-0.73

iesel

-0.72

ython

-0.71

daily

-0.71

insured

-0.70

 Seym

-0.70

POSITIVE LOGITS

hunt

1.22

hood

1.13

gling

0.98

uscript

0.94

volent

0.92

liness

0.90

agers

0.90

ifest

0.89

liest

0.88

nered

0.81

Activations Density 0.061%

proper nouns or titles related to individuals or characters

mentions of the word "man"

No Comments

No Known Activations