INDEX

Explanations

references to traps

oai_token-act-pair · gpt-3.5-turbo

references to various types of traps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

edly

-0.75

issance

-0.73

sis

-0.73

 normalized

-0.67

league

-0.66

 Kardashian

-0.66

Uz

-0.63

played

-0.63

Merit

-0.63

kees

-0.62

POSITIVE LOGITS

door

1.50

doors

1.35

trap

1.01

finding

0.97

 traps

0.90

idon

0.87

 trap

0.84

ERY

0.80

 Trap

0.78

 Traps

0.75

Activations Density 0.023%

references to traps

references to various types of traps

No Comments

No Known Activations