INDEX

Explanations

instances where something has reached a particular level or point

oai_token-act-pair · gpt-3.5-turbo

references to levels of achievement or milestones

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

serv

-0.77

ItemImage

-0.76

 faithfully

-0.75

 followed

-0.74

 alone

-0.72

CT

-0.68

 accompanied

-0.66

ItemThumbnailImage

-0.65

.]

-0.65

 accordingly

-0.64

POSITIVE LOGITS

 brink

1.34

 pinnacle

1.30

 point

1.23

 conclusion

1.17

 tipping

1.14

 verge

1.14

 limit

1.12

end

1.11

 threshold

1.11

 halfway

1.08

Activations Density 0.140%

instances where something has reached a particular level or point

references to levels of achievement or milestones

No Comments

No Known Activations