INDEX

Explanations

lines of code containing specific instructions for Intel x86 architecture

oai_token-act-pair · gpt-3.5-turbo

references to the x86 architecture in computing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

vati

-0.73

jriwal

-0.70

pressure

-0.68

 infl

-0.67

*/(

-0.66

uyomi

-0.66

 agitated

-0.66

itton

-0.65

eday

-0.65

igham

-0.65

POSITIVE LOGITS

âĶĢ

0.88

%:

0.83

0.79

0.75

URRENT

0.74

ENTION

0.74

0.73

 Minutes

0.72

0.70

Activations Density 0.029%

lines of code containing specific instructions for Intel x86 architecture

references to the x86 architecture in computing

No Comments

No Known Activations