INDEX

Explanations

references to specific buildings and structures, particularly those related to the World Trade Center

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nir

-0.88

xon

-0.82

romeda

-0.82

ican

-0.81

sha

-0.81

flix

-0.80

opard

-0.79

ricular

-0.75

icans

-0.72

 Nadu

-0.71

POSITIVE LOGITS

 renovations

1.00

 blocks

1.00

 occupancy

0.99

raper

0.98

Constructed

0.98

 erected

0.94

 occupants

0.93

 renovation

0.92

 constructed

0.91

 demolition

0.90

Activations Density 0.038%

references to specific buildings and structures, particularly those related to the World Trade Center

No Comments

No Known Activations