INDEX

Explanations

medical terms related to surgery and surgeons

oai_token-act-pair · gpt-3.5-turbo

references to surgeons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

nd

-0.70

atives

-0.66

onies

-0.64

EVs

-0.61

cks

-0.59

formance

-0.59

-|

-0.58

 Prot

-0.58

redits

-0.58

ks

-0.58

POSITIVE LOGITS

 surgeon

3.85

 surgeons

3.23

 Surgery

1.96

 surgery

1.75

 physician

1.65

 surgical

1.65

 psychiatrist

1.58

 veterinarian

1.56

 dentist

1.55

 therapist

1.49

Activations Density 0.008%

medical terms related to surgery and surgeons

references to surgeons

No Comments

No Known Activations