INDEX

Explanations

references to specific geographic locations, particularly cities and regions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

ãĤ¯

-0.72

Ø¨

-0.68

assad

-0.67

Ott

-0.66

inav

-0.65

 favour

-0.65

anted

-0.65

comfort

-0.65

rav

-0.64

edition

-0.63

POSITIVE LOGITS

 County

1.26

 Mountains

1.05

 Island

1.05

 Valley

1.00

 Unified

0.96

 Fault

0.92

 Islands

0.91

 Rapids

0.90

 Province

0.90

 Springs

0.87

Activations Density 0.023%

references to specific geographic locations, particularly cities and regions

No Comments

No Known Activations