INDEX

Explanations

capital letters 'C' followed by a single digit

oai_token-act-pair · gpt-3.5-turbo

references to the letter 'C'

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥĦ

-0.85

 GOODMAN

-0.66

hof

-0.66

ãĤ¤ãĥĪ

-0.66

ItemTracker

-0.65

ãĥĹ

-0.65

 absor

-0.64

ãĤ¦ãĤ¹

-0.64

EStream

-0.63

Magikarp

-0.63

POSITIVE LOGITS

overed

1.08

ancel

1.05

ursor

1.02

argo

1.01

ASE

0.98

olly

0.98

ortex

0.97

ologne

0.97

URRENT

0.97

elsius

0.97

Activations Density 0.051%

capital letters 'C' followed by a single digit

references to the letter 'C'

No Comments

No Known Activations