INDEX

Explanations

questions starting with "What does" in the context of seeking meaning or implications

oai_token-act-pair · gpt-3.5-turbo

questions and references to meaning or implications

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

geoning

-0.71

TPPStreamerBot

-0.71

imov

-0.70

oys

-0.63

sung

-0.63

noticed

-0.62

oved

-0.62

geon

-0.62

abad

-0.62

mort

-0.61

POSITIVE LOGITS

 entail

1.38

 entails

1.33

 mean

1.25

 Means

1.12

 means

1.07

 meant

1.05

mean

1.00

 signify

1.00

 signifies

0.93

 boils

0.90

Activations Density 0.129%

questions starting with "What does" in the context of seeking meaning or implications

questions and references to meaning or implications

No Comments

No Known Activations