INDEX

Explanations

references to the name "Horn."

oai_token-act-pair · gpt-3.5-turbo

references to "horn" or "horns."

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

NCT

-0.93

arnaev

-0.74

PsyNetMessage

-0.67

ufact

-0.67

FORM

-0.66

juries

-0.62

 Chatt

-0.61

ENCY

-0.60

 Scotia

-0.60

POSITIVE LOGITS

indal

0.99

pipe

0.98

worms

0.91

sey

0.91

wig

0.86

atility

0.84

 Horn

0.84

worm

0.83

beck

0.83

bill

0.83

Activations Density 0.016%

references to the name "Horn."

references to "horn" or "horns."

No Comments

No Known Activations