INDEX

Explanations

phrases related to finality or conclusion

oai_token-act-pair · gpt-3.5-turbo

phrases and references to the concept of being "last" or "the final" in various contexts

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

obin

-0.80

agar

-0.65

olves

-0.64

also

-0.63

hovah

-0.63

went

-0.62

ingen

-0.61

ahon

-0.61

olor

-0.61

rovers

-0.61

POSITIVE LOGITS

 gasp

1.31

 remnants

1.16

 rites

1.12

 straw

1.12

 remaining

1.10

 remnant

1.09

 vest

1.09

 bast

1.04

 ditch

1.03

 surviving

1.01

Activations Density 0.064%

phrases related to finality or conclusion

phrases and references to the concept of being "last" or "the final" in various contexts

No Comments

No Known Activations