INDEX

Explanations

references to the location "Pearl Harbor."

oai_token-act-pair · gpt-3.5-turbo

references to Pearl Harbor

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

enegger

-0.88

NRS

-0.85

dit

-0.79

SPONSORED

-0.78

ioch

-0.73

hered

-0.73

ictions

-0.70

 prevailing

-0.67

 therap

-0.66

elsius

-0.66

POSITIVE LOGITS

 Harbor

1.15

 Pearl

1.08

ridge

0.96

 Harbour

0.92

stein

0.92

ãĥĥãĥī

0.91

Pear

0.91

sburg

0.79

fish

0.75

 Flask

0.74

Activations Density 0.009%

references to the location "Pearl Harbor."

references to Pearl Harbor

No Comments

No Known Activations