INDEX

Explanations

phrases related to traps

oai_token-act-pair · gpt-3.5-turbo

references to traps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

issance

-0.94

edly

-0.93

sis

-0.74

iations

-0.68

ial

-0.67

hip

-0.62

rates

-0.62

care

-0.60

ials

-0.59

cox

-0.59

POSITIVE LOGITS

door

1.52

doors

1.40

 traps

0.85

finding

0.83

trap

0.83

idon

0.82

 trap

0.80

 lure

0.79

 bait

0.74

 Hole

0.72

Activations Density 0.045%

phrases related to traps

references to traps

No Comments

No Known Activations