INDEX

Explanations

phrases related to technical commands or instructions

oai_token-act-pair · gpt-3.5-turbo

punctuation and formatting variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ariat

-0.66

Yon

-0.64

 Reloaded

-0.63

 Hague

-0.62

 Aden

-0.60

 Canaver

-0.58

VID

-0.57

${

-0.56

apego

-0.55

çĦ

-0.55

POSITIVE LOGITS

 alike

0.74

 sensibilities

0.72

Ħ¢

0.71

 environments

0.67

ĨĴ

0.66

 respectively

0.65

ngth

0.64

missions

0.64

cation

0.63

 Dragonbound

0.62

Activations Density 0.979%

phrases related to technical commands or instructions

punctuation and formatting variations

No Comments

No Known Activations