INDEX

Explanations

words related to security or protection

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of "poaching."

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

owship

-0.86

hips

-0.74

ATIONAL

-0.72

IELD

-0.72

ITNESS

-0.68

NESS

-0.67

 REAL

-0.65

chell

-0.65

 Catalyst

-0.64

EMENT

-0.63

POSITIVE LOGITS

pper

1.03

etary

0.92

oper

0.92

achers

0.91

opy

0.90

pping

0.89

ppy

0.89

oting

0.88

Äį

0.88

pert

0.87

Activations Density 0.010%

words related to security or protection

words related to the concept of "poaching."

No Comments

No Known Activations