INDEX

Explanations

names, specifically those related to French and Spanish individuals

oai_token-act-pair · gpt-3.5-turbo

names of individuals or references to people

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

iary

-0.89

oky

-0.82

-0.80

ipt

-0.78

ividual

-0.76

iqueness

-0.75

oodle

-0.74

ramid

-0.74

dream

-0.74

iating

-0.73

POSITIVE LOGITS

 Hollande

1.06

 Franco

0.99

 Francois

0.90

 Dres

0.88

 fries

0.80

ugal

0.77

 FranÃ§ois

0.76

bourg

0.75

olini

0.72

furt

0.72

Activations Density 0.021%

names, specifically those related to French and Spanish individuals

names of individuals or references to people

No Comments

No Known Activations