INDEX

Explanations

terms related to the concept of 'ham' or 'hamstring'

oai_token-act-pair · gpt-3.5-turbo

mentions of "ham" and its related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Occupations

-0.70

URES

-0.66

 Statements

-0.65

URE

-0.64

 Holder

-0.62

tical

-0.61

 Appropri

-0.60

 Cardinal

-0.60

 Scrolls

-0.59

aunder

-0.59

POSITIVE LOGITS

ham

1.20

ilton

1.13

strings

1.12

gob

1.05

elin

1.05

pering

0.96

stead

0.91

iago

0.89

mers

0.88

wich

0.86

Activations Density 0.004%

terms related to the concept of 'ham' or 'hamstring'

mentions of "ham" and its related terms

No Comments

No Known Activations