INDEX

Explanations

keywords related to user experiences and interactions

oai_token-act-pair · gpt-3.5-turbo

mentions of user experiences in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fam

-0.69

laws

-0.69

hatt

-0.69

ependent

-0.68

law

-0.68

vous

-0.65

 actionGroup

-0.64

roup

-0.64

 stockp

-0.63

 clot

-0.63

POSITIVE LOGITS

 Experience

1.05

Experience

1.01

 experience

0.93

iences

0.92

 experiences

0.90

Capture

0.83

ually

0.80

IENCE

0.80

ttes

0.79

ience

0.79

Activations Density 0.024%

keywords related to user experiences and interactions

mentions of user experiences in various contexts

No Comments

No Known Activations