INDEX

Explanations

mentions or appearances of the abbreviation "sf"

oai_token-act-pair · gpt-3.5-turbo

terms related to substantive content and specific entities or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ModLoader

-0.86

©¶æ

-0.71

 Staten

-0.69

 destro

-0.69

PATH

-0.69

Reloaded

-0.66

xtap

-0.63

geons

-0.61

 Myster

-0.60

XL

-0.60

POSITIVE LOGITS

emp

0.78

elvet

0.77

assian

0.77

arian

0.73

apolis

0.72

ixed

0.72

agar

0.72

itude

0.69

inion

0.69

enton

0.69

Activations Density 0.022%

mentions or appearances of the abbreviation "sf"

terms related to substantive content and specific entities or organizations

No Comments

No Known Activations