INDEX

Explanations

websites or domain names

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of a specific term or acronym, particularly "WA"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Scient

-0.69

 forth

-0.69

 holog

-0.67

 Malt

-0.67

 dialog

-0.65

rou

-0.65

 Proced

-0.63

 Cour

-0.63

 Czech

-0.63

 augmented

-0.61

POSITIVE LOGITS

WA

4.21

WC

1.43

WR

1.37

WI

1.35

wa

1.25

FA

1.19

HA

1.18

PA

1.18

SA

1.17

GA

1.17

Activations Density 0.017%

websites or domain names

repeated mentions of a specific term or acronym, particularly "WA"

No Comments

No Known Activations