INDEX

Explanations

terms related to capitalization

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Cap" and its variations, likely related to capitalization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

hower

-0.96

ĪĴ

-0.85

 silence

-0.78

¿½

-0.76

 anten

-0.68

hood

-0.66

perse

-0.65

cause

-0.64

IGHTS

-0.63

selves

-0.63

POSITIVE LOGITS

itol

1.40

acity

1.28

abilities

1.28

itals

1.26

illary

1.21

rice

1.16

rices

1.11

uchin

1.10

itan

1.07

asso

1.02

Activations Density 0.020%

terms related to capitalization

occurrences of the word "Cap" and its variations, likely related to capitalization

No Comments

No Known Activations