INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 dated

-0.67

ccording

-0.66

 helped

-0.63

-0.60

 aided

-0.59

 teamed

-0.57

 backed

-0.56

 accompanied

-0.56

çīĪ

-0.56

 robotic

-0.55

POSITIVE LOGITS

um

2.49

anus

1.06

orum

0.94

ums

0.93

UM

0.90

ume

0.84

uman

0.82

uri

0.81

aria

0.77

avia

0.76

Activations Density 0.000%

No Known Activations

This feature has no known activations.