INDEX

Explanations

years of events as noted in citations

oai_token-act-pair · gpt-3.5-turbo

commas and other punctuation marks in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 intensive

-0.69

aukee

-0.66

nesday

-0.66

uters

-0.66

 manufact

-0.64

 mosqu

-0.63

ateurs

-0.63

 reinvest

-0.62

rikes

-0.61

 UNCLASSIFIED

-0.61

POSITIVE LOGITS

Pg

1.24

src

1.15

note

1.13

...]

1.09

 citation

1.01

?]

0.99

Footnote

0.98

âĢ¦]

0.94

ËĪ

0.93

sic

0.92

Activations Density 0.036%

years of events as noted in citations

commas and other punctuation marks in the text

No Comments

No Known Activations