INDEX

Explanations

locations and terms related to places, particularly in the context of conflict, captivity, or emergency situations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

thood

-0.86

OME

-0.66

wcsstore

-0.66

imon

-0.65

haar

-0.64

ategor

-0.63

cale

-0.63

powers

-0.63

onnaissance

-0.63

ãĥīãĥ©

-0.62

POSITIVE LOGITS

 illegally

0.86

 whence

0.86

 precinct

0.84

 vicinity

0.81

 perimeter

0.79

 shortly

0.79

 courthouse

0.75

 itself

0.75

 situated

0.73

 premises

0.73

Activations Density 0.132%

locations and terms related to places, particularly in the context of conflict, captivity, or emergency situations

No Comments

No Known Activations