INDEX

Explanations

mentions of award-winning achievements

oai_token-act-pair · gpt-3.5-turbo

references to award-winning achievements

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

omach

-0.89

alle

-0.83

trak

-0.82

ugu

-0.80

otor

-0.77

agen

-0.75

agan

-0.74

tera

-0.74

ase

-0.71

gun

-0.69

POSITIVE LOGITS

fare

0.74

Initialized

0.71

theless

0.70

 laureate

0.69

ãĤ¦ãĤ¹

0.68

 Pulitzer

0.68

NESS

0.67

 Winston

0.65

 BaseType

0.65

AFTA

0.64

Activations Density 0.022%

mentions of award-winning achievements

references to award-winning achievements

No Comments

No Known Activations