INDEX

Explanations

phrases related to various usernames or profiles on websites

oai_token-act-pair · gpt-3.5-turbo

entities or usernames associated with online platforms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 readiness

-0.86

 proble

-0.85

 UNESCO

-0.83

 authenticated

-0.80

 succession

-0.79

 âĢ¦"

-0.79

 broch

-0.78

 depreciation

-0.77

 conduc

-0.76

 unaccompanied

-0.76

POSITIVE LOGITS

Gaming

1.14

gamer

1.14

Gamer

1.12

Wolf

1.11

Games

1.04

bear

1.04

monkey

1.04

Knight

1.01

gaming

1.01

Nin

0.97

Activations Density 0.464%

phrases related to various usernames or profiles on websites

entities or usernames associated with online platforms

No Comments

No Known Activations