INDEX

Explanations

questions starting with "Do" and followed by different clauses or subjects

oai_token-act-pair · gpt-3.5-turbo

questions starting with the word "Do."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 accrued

-0.66

 surfaced

-0.63

 restored

-0.63

 dimin

-0.63

 emerged

-0.62

pal

-0.62

 circulation

-0.60

 punches

-0.60

 circulating

-0.60

 available

-0.60

POSITIVE LOGITS

Do

3.29

Do

2.03

DO

1.92

do

1.92

Are

1.71

Have

1.68

Did

1.63

Does

1.60

Don

1.56

Would

1.41

Activations Density 0.006%

questions starting with "Do" and followed by different clauses or subjects

questions starting with the word "Do."

No Comments

No Known Activations