INDEX

Explanations

numbers, codes, and technical details

oai_token-act-pair · gpt-3.5-turbo

quantities and measurements related to various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

stood

-0.86

eworld

-0.71

ews

-0.69

assetsadobe

-0.69

eworks

-0.69

 runaway

-0.68

-0.67

 Daesh

-0.61

rontal

-0.61

izoph

-0.60

POSITIVE LOGITS

2.93

2.58

**

2.47

*)

2.35

*.

2.27

*,

2.27

)*

2.24

.*

2.21

*/

2.18

*:

2.11

Activations Density 0.192%

numbers, codes, and technical details

quantities and measurements related to various contexts

No Comments

No Known Activations