INDEX

Explanations

phrases related to medical symptoms and conditions

oai_token-act-pair · gpt-3.5-turbo

references to symptoms associated with medical conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

cean

-0.72

 holdings

-0.69

adr

-0.69

 Expend

-0.69

ramid

-0.66

 spons

-0.63

 sanct

-0.62

ilings

-0.62

 GEAR

-0.62

 ownership

-0.62

POSITIVE LOGITS

ptoms

1.44

 symptoms

1.36

 Symptoms

1.32

 symptom

1.10

 suggestive

1.07

 worsen

1.05

 onset

0.96

 worsened

0.90

 worsening

0.90

 resembling

0.87

Activations Density 0.082%

phrases related to medical symptoms and conditions

references to symptoms associated with medical conditions

No Comments

No Known Activations