INDEX

Explanations

phrases related to testing, effectiveness, and deployment of various methods and technologies

oai_token-act-pair · gpt-3.5-turbo

phrases related to the effectiveness and testing of treatments or therapies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cknowled

-0.78

 largeDownload

-0.74

 heels

-0.69

Tickets

-0.67

 Stein

-0.65

DragonMagazine

-0.65

 Credits

-0.64

 Fein

-0.64

IMF

-0.61

Optional

-0.60

POSITIVE LOGITS

acea

0.93

humans

0.79

 humans

0.74

chnology

0.74

 interstate

0.73

 everyday

0.70

 therape

0.69

 throughput

0.69

anced

0.68

 harming

0.66

Activations Density 0.515%

phrases related to testing, effectiveness, and deployment of various methods and technologies

phrases related to the effectiveness and testing of treatments or therapies

No Comments

No Known Activations