INDEX

Explanations

expressions related to computer architecture, particularly x86

oai_token-act-pair · gpt-3.5-turbo

references to the x86 computer architecture and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 slit

-0.77

*/(

-0.73

 compr

-0.71

RTX

-0.70

pees

-0.67

 ascent

-0.67

apo

-0.65

awed

-0.63

 blurred

-0.63

alid

-0.63

POSITIVE LOGITS

âĶĢ

0.81

0.80

lli

0.78

%:

0.77

xual

0.72

Ñĭ

0.70

chan

0.69

rpm

0.69

ï¿½ï¿½ï¿½ï¿½

0.68

OTAL

0.67

Activations Density 0.040%

expressions related to computer architecture, particularly x86

references to the x86 computer architecture and its variants

No Comments

No Known Activations