INDEX

Explanations

specific code snippets related to manipulating user data

oai_token-act-pair · gpt-3.5-turbo

identifiers or keys in a coding or database context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

psey

-0.84

avorite

-0.83

dinand

-0.83

everal

-0.83

ĸļ

-0.83

regor

-0.82

ixel

-0.82

atre

-0.82

¥ŀ

-0.81

 behavi

-0.80

POSITIVE LOGITS

 undefined

0.95

nil

0.91

 Unknown

0.87

hover

0.86

 Logged

0.85

*/

0.80

{}

0.80

+#

0.79

",

0.78

 unspecified

0.77

Activations Density 0.101%

specific code snippets related to manipulating user data

identifiers or keys in a coding or database context

No Comments

No Known Activations