INDEX

Explanations

phrases that end with a particular symbol or word

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and ellipses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ãĥ¼ãĥĨãĤ£

-1.06

ãĥ¼ãĥĨ

-0.81

ãĥ³ãĤ¸

-0.74

nen

-0.73

ãĥ«

-0.71

ocratic

-0.70

ãĥİ

-0.70

iled

-0.69

emia

-0.66

orah

-0.66

POSITIVE LOGITS

...]

0.81

mmmm

0.80

jpg

0.79

Pg

0.78

picture

0.76

.....

0.75

....

0.74

âĢ¦]

0.74

pict

0.73

 Personally

0.73

Activations Density 0.012%

phrases that end with a particular symbol or word

punctuation marks and ellipses

No Comments

No Known Activations