INDEX

Explanations

references to experience levels in relation to learning specific skills or tools

oai_token-act-pair · gpt-3.5-turbo

references to beginners and entry-level individuals in the context of learning or engagement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

osate

-0.87

 disputed

-0.72

 è£ıç

-0.71

itol

-0.70

rocal

-0.70

Closure

-0.67

imony

-0.67

ossession

-0.66

trak

-0.65

ess

-0.65

POSITIVE LOGITS

 beginners

1.78

 beginner

1.60

 novice

1.56

 newcomers

1.54

 enthusiasts

1.50

 professionals

1.40

 gamers

1.36

 learners

1.33

 hobby

1.33

 nerds

1.31

Activations Density 0.404%

references to experience levels in relation to learning specific skills or tools

references to beginners and entry-level individuals in the context of learning or engagement

No Comments

No Known Activations