INDEX

Explanations

words related to geographical locations, specifically focusing on the word "Prairie"

oai_token-act-pair · gpt-3.5-turbo

references to prairies and plains

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

ISTER

-0.80

ocker

-0.78

isters

-0.76

igmat

-0.74

iction

-0.73

arov

-0.73

undo

-0.71

istration

-0.71

andowski

-0.71

ITED

-0.71

POSITIVE LOGITS

irie

1.00

 Prairie

0.86

 Trace

0.74

 Delta

0.71

ĺħ

0.70

brush

0.69

 Amar

0.68

 Springs

0.68

 Vaugh

0.67

tti

0.66

Activations Density 0.025%

words related to geographical locations, specifically focusing on the word "Prairie"

references to prairies and plains

No Comments

No Known Activations