INDEX

Explanations

descriptions of individuals or characters, including their roles and associations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the" 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 owing

-0.78

fax

-0.78

 complying

-0.77

hops

-0.74

strap

-0.74

 accordingly

-0.74

merce

-0.73

.--

-0.72

zon

-0.72

.-

-0.70

POSITIVE LOGITS

 infamous

1.13

 Clintons

1.07

 aforementioned

1.04

ocratic

1.03

 famed

1.03

 same

1.03

 legendary

1.00

 latter

0.98

 youngest

0.98

 dreaded

0.97

Activations Density 0.911%

descriptions of individuals or characters, including their roles and associations

occurrences of the word "the" in various contexts

No Comments

No Known Activations