INDEX

Explanations

keywords related to cycling and cyclists

oai_token-act-pair · gpt-3.5-turbo

terms related to cycling and cyclists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

arial

-0.95

urities

-0.81

IELD

-0.77

nesses

-0.75

orial

-0.75

aples

-0.74

enger

-0.73

urated

-0.71

utterstock

-0.71

dated

-0.70

POSITIVE LOGITS

cling

1.00

 cyclist

1.00

 cycling

0.91

 cyclists

0.88

puter

0.83

 Cycling

0.79

 riders

0.78

Tips

0.78

erg

0.78

 borne

0.77

Activations Density 0.011%

keywords related to cycling and cyclists

terms related to cycling and cyclists

No Comments

No Known Activations