INDEX

Explanations

the word "hunt."

oai_token-act-pair · gpt-3.5-turbo

terms related to searches or pursuits, particularly in a context of investigation or hunting

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

 bottom

-0.70

 correctly

-0.66

 Democr

-0.65

 wrong

-0.64

 correct

-0.64

 joints

-0.64

 receiving

-0.64

 properly

-0.63

 consent

-0.62

ink

-0.62

POSITIVE LOGITS

hunt

4.17

hun

3.25

hunter

2.37

Hunt

1.38

Hun

1.30

hound

1.28

lihood

1.11

har

1.07

haul

1.06

catch

1.02

Activations Density 0.006%

the word "hunt."

terms related to searches or pursuits, particularly in a context of investigation or hunting

No Comments

No Known Activations