INDEX

Explanations

companies or organizations

oai_token-act-pair · gpt-3.5-turbo

comparisons involving the word "like."

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

Published

-0.76

hiba

-0.72

inas

-0.71

zees

-0.70

erial

-0.69

ells

-0.67

atched

-0.67

 showc

-0.67

ulty

-0.66

ipple

-0.66

POSITIVE LOGITS

lihood

1.55

lier

1.10

 minded

0.97

liest

0.95

minded

0.92

 ours

0.90

 yours

0.79

liness

0.78

 wildfire

0.77

 theirs

0.73

Activations Density 0.069%

companies or organizations

comparisons involving the word "like."

No Comments

No Known Activations