INDEX

Explanations

text excerpts

oai_token-act-pair · gpt-3.5-turbo

references to excerpts and snippets from documents

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

ggle

-1.01

gaard

-0.79

thood

-0.74

mos

-0.71

ggles

-0.69

kos

-0.69

dream

-0.68

ceans

-0.67

Â¯Â¯Â¯Â¯

-0.67

vil

-0.64

POSITIVE LOGITS

 excerpts

1.21

 excerpt

1.13

 snippets

1.00

 guiActiveUn

0.97

cerpt

0.93

DragonMagazine

0.93

 extracts

0.83

icularly

0.82

 summar

0.81

 Transcript

0.81

Activations Density 0.017%

text excerpts

references to excerpts and snippets from documents

No Comments

No Known Activations