INDEX

Explanations

medical terms related to blood flow

oai_token-act-pair · gpt-3.5-turbo

terms related to measurements and maximum values

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

trak

-0.83

 pert

-0.67

 factor

-0.66

 body

-0.66

 swallow

-0.66

asc

-0.62

Var

-0.62

lie

-0.61

 trait

-0.60

body

-0.60

POSITIVE LOGITS

 Juda

0.83

ABE

0.83

FTWARE

0.77

 Puzzles

0.74

 Azure

0.74

thumbnails

0.70

 Spaces

0.68

 CODE

0.67

 Warehouse

0.66

ENTS

0.66

Activations Density 0.000%

medical terms related to blood flow

terms related to measurements and maximum values

No Comments

No Known Activations