INDEX

Explanations

mentions of positive reviews on Amazon

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "on."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

auga

-0.70

eties

-0.67

mere

-0.66

EngineDebug

-0.65

arnaev

-0.65

htaking

-0.64

Â¯Â¯

-0.64

ptive

-0.62

lethal

-0.62

amount

-0.62

POSITIVE LOGITS

 reddit

1.45

 facebook

1.44

 youtube

1.41

 Youtube

1.38

 Reddit

1.35

 twitter

1.34

 behalf

1.32

 Github

1.30

 forums

1.25

 Facebook

1.25

Activations Density 0.160%

mentions of positive reviews on Amazon

occurrences of the word "on."

No Comments

No Known Activations