INDEX

Explanations

documents or information that are related to a specific topic or subject

oai_token-act-pair · gpt-3.5-turbo

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

ierra

-0.59

uckland

-0.57

oker

-0.57

owler

-0.57

é¾įå

-0.56

ONES

-0.56

imates

-0.55

âķĲ

-0.55

lite

-0.55

ãĤ©

-0.55

POSITIVE LOGITS

 thereto

1.44

to

1.12

 specifically

0.83

 solely

0.77

to

0.74

 directly

0.72

 unto

0.72

 exclusively

0.71

 principally

0.70

 favorably

0.70

Activations Density 0.057%

documents or information that are related to a specific topic or subject

No Comments

No Known Activations