INDEX

Explanations

references to research studies or reports

oai_token-act-pair · gpt-3.5-turbo

references to new studies and reports

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

artifacts

-0.65

 Machines

-0.63

animate

-0.63

headers

-0.63

 Himself

-0.62

 Ajax

-0.62

 Blades

-0.60

 Horus

-0.60

lihood

-0.59

 mortals

-0.58

POSITIVE LOGITS

 titled

0.85

 entitled

0.77

 headlined

0.70

 commissioned

0.69

neum

0.68

lished

0.67

 published

0.66

 reviewer

0.66

itled

0.64

 researcher

0.63

Activations Density 0.197%

references to research studies or reports

references to new studies and reports

No Comments

No Known Activations