INDEX

Explanations

mentions of specific events or locations

oai_token-act-pair · gpt-3.5-turbo

specific terms related to databases and related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hal

-0.74

ãĥŁ

-0.72

illac

-0.66

ãĥķãĤ©

-0.66

hatt

-0.65

ranch

-0.65

":["

-0.64

ï¸ı

-0.63

hab

-0.61

POSE

-0.61

POSITIVE LOGITS

by

1.78

by

1.78

By

1.38

BY

1.33

bys

1.33

By

1.31

BY

1.13

ById

0.89

byn

0.89

via

0.80

Activations Density 0.216%

mentions of specific events or locations

specific terms related to databases and related activities

No Comments

No Known Activations