INDEX

Explanations

the word "newly" followed by a number

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "newly."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ashtra

-0.79

trak

-0.78

orem

-0.78

rice

-0.76

ospace

-0.75

ocene

-0.72

 Bastard

-0.72

osate

-0.68

 Citation

-0.66

ertodd

-0.66

POSITIVE LOGITS

ãĤ»

0.95

ãĤ´

0.89

bie

0.84

foundland

0.82

bies

0.82

wed

0.82

 discovered

0.79

 reunited

0.77

ESH

0.76

 appointed

0.73

Activations Density 0.007%

the word "newly" followed by a number

occurrences of the word "newly."

No Comments

No Known Activations