INDEX

Explanations

places and landscapes related to mountains

oai_token-act-pair · gpt-3.5-turbo

references to mountainous locations

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

tle

-0.75

NER

-0.70

INAL

-0.70

IAL

-0.69

ABLE

-0.67

andum

-0.64

ASED

-0.62

ership

-0.62

 absentee

-0.60

phas

-0.60

POSITIVE LOGITS

mith

1.02

hooting

1.01

ides

0.99

creen

0.98

cape

0.95

 challeng

0.89

hips

0.88

ight

0.87

hed

0.87

cale

0.86

Activations Density 0.052%

places and landscapes related to mountains

references to mountainous locations

No Comments

No Known Activations