INDEX

Explanations

dates and chronological information

oai_token-act-pair · gpt-3.5-turbo

dates and numerical information related to time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Burk

-0.85

Faw

-0.85

 Comet

-0.82

buck

-0.82

Poc

-0.80

éĥ

-0.80

ados

-0.79

 Conway

-0.79

 Hawth

-0.77

oust

-0.76

POSITIVE LOGITS

1.28

1.27

XII

0.91

0.87

0.86

0.80

LR

0.80

Rah

0.79

 Martinez

0.79

Activations Density 0.334%

dates and chronological information

dates and numerical information related to time

No Comments

No Known Activations