INDEX

Explanations

phrases related to ongoing discussions or interviews

oai_token-act-pair · gpt-3.5-turbo

mentions of ongoing discussions or interviews

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 transferring

-0.68

 valves

-0.66

 signaling

-0.65

alias

-0.64

voc

-0.64

License

-0.64

alid

-0.64

entially

-0.64

alties

-0.63

RIC

-0.63

POSITIVE LOGITS

 roundup

1.30

 podcast

1.09

 interview

1.06

 rundown

1.01

 recap

1.00

 Podcast

1.00

 slideshow

0.98

 coverage

0.98

Ask

0.96

 preview

0.96

Activations Density 0.283%

phrases related to ongoing discussions or interviews

mentions of ongoing discussions or interviews

No Comments

No Known Activations