INDEX

Explanations

references to a specific location named "York" with varying specificity

oai_token-act-pair · gpt-3.5-turbo

mentions of "York" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

cius

-0.82

aeda

-0.77

Ã±a

-0.75

itures

-0.73

ebus

-0.70

Ã±

-0.67

 subsistence

-0.66

 careful

-0.66

Ú

-0.64

cedes

-0.63

POSITIVE LOGITS

mares

1.07

dale

1.01

mont

0.86

town

0.77

bean

0.76

onne

0.76

onge

0.76

stones

0.76

York

0.74

burgh

0.74

Activations Density 0.016%

references to a specific location named "York" with varying specificity

mentions of "York" in various contexts

No Comments

No Known Activations