INDEX

Explanations

additional information or resources related to the main content

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "Further Reading."

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

 smile

-0.75

 usually

-0.70

 practiced

-0.69

win

-0.67

 smiles

-0.67

 wins

-0.65

aten

-0.65

 instinct

-0.64

 clichÃ©

-0.63

old

-0.62

POSITIVE LOGITS

Further

3.37

 Further

2.57

urther

1.63

Furthermore

1.48

 further

1.38

Additional

1.31

Moreover

1.31

Additionally

1.29

Secondly

1.27

Beyond

1.26

Activations Density 0.012%

additional information or resources related to the main content

occurrences of the phrase "Further Reading."

No Comments

No Known Activations