INDEX

Explanations

sources or attributions within text

oai_token-act-pair · gpt-3.5-turbo

references to sources or citations

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

estern

-0.90

uckle

-0.78

 tremend

-0.76

 sled

-0.76

 teasp

-0.74

 metic

-0.74

vow

-0.73

 sway

-0.72

 stomp

-0.71

 exting

-0.70

POSITIVE LOGITS

Source

1.25

 Sources

1.24

source

1.17

 Source

1.15

Sources

1.13

ource

1.11

 Citation

0.88

Fed

0.88

ources

0.84

Cub

0.82

Activations Density 0.008%

sources or attributions within text

references to sources or citations

No Comments

No Known Activations