INDEX

Explanations

phrases related to research, analysis, investigation, and assessment

oai_token-act-pair · gpt-3.5-turbo

phrases indicating various types of assessments, inspections, and evaluations

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

osponsors

-0.79

 Survivors

-0.69

 Fried

-0.66

 Sett

-0.66

 Achievements

-0.65

 Tooth

-0.62

mates

-0.62

 Hitch

-0.59

 shores

-0.59

dt

-0.59

POSITIVE LOGITS

 utmost

0.82

MRI

0.81

 Advertisement

0.80

 diligence

0.76

 captcha

0.71

ãĥĨ

0.71

auntlet

0.70

isition

0.69

issance

0.69

cule

0.67

Activations Density 0.245%

phrases related to research, analysis, investigation, and assessment

phrases indicating various types of assessments, inspections, and evaluations

No Comments

No Known Activations