INDEX

Explanations

references to entities named "Glen"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Glen" and "Monte"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

enance

-0.71

į

-0.66

entric

-0.65

 ransom

-0.64

 diaper

-0.64

hips

-0.64

 Mehran

-0.61

 lockout

-0.60

 Haram

-0.59

 scraps

-0.59

POSITIVE LOGITS

coe

1.31

cone

1.00

forth

0.98

apple

0.97

felt

0.96

vir

0.95

voy

0.93

ritional

0.92

gow

0.92

ovie

0.91

Activations Density 0.049%

references to entities named "Glen"

mentions of the name "Glen" and "Monte"

No Comments

No Known Activations