INDEX

Explanations

mentions of objects or concepts related to concealment or mystery, such as veils, cloaks, and shrouds

oai_token-act-pair · gpt-3.5-turbo

terms related to concealment or secrecy

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

£ı

-0.73

thritis

-0.72

Unit

-0.72

olesc

-0.72

ymes

-0.71

aline

-0.71

 Stim

-0.71

pson

-0.69

itialized

-0.69

ellig

-0.67

POSITIVE LOGITS

 shroud

1.30

 veil

1.19

 cloak

0.96

 veiled

0.90

 maiden

0.86

fold

0.86

 Veil

0.86

 draped

0.85

 curtain

0.83

 gown

0.81

Activations Density 0.018%

mentions of objects or concepts related to concealment or mystery, such as veils, cloaks, and shrouds

terms related to concealment or secrecy

No Comments

No Known Activations