INDEX

Explanations

clear and precise descriptions or statements

oai_token-act-pair · gpt-3.5-turbo

instances of the word "clear" and related phrases emphasizing clarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

FactoryReloaded

-0.70

 awkwardly

-0.66

 passively

-0.64

 tremend

-0.63

ãĥĦ

-0.62

Meh

-0.61

 Loft

-0.61

uld

-0.61

 temporarily

-0.60

 McGu

-0.60

POSITIVE LOGITS

cut

1.39

ances

1.34

cutting

1.09

headed

1.08

 indication

1.02

 deline

1.01

cuts

1.01

ance

0.91

 winner

0.86

seeing

0.85

Activations Density 0.061%

clear and precise descriptions or statements

instances of the word "clear" and related phrases emphasizing clarity

No Comments

No Known Activations