INDEX

Explanations

mentions of locations and actions related to law enforcement or criminal incidents

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

 compounded

-0.66

 transmitted

-0.65

 attracted

-0.64

 compared

-0.63

 hurled

-0.62

 preceded

-0.62

 complied

-0.62

 graz

-0.62

 employed

-0.61

 phased

-0.61

POSITIVE LOGITS

pless

1.01

 grips

0.89

asted

0.86

ggles

0.85

 extremes

0.80

bed

0.79

iling

0.77

 meet

0.74

 Dover

0.73

 relieve

0.72

Activations Density 0.850%

mentions of locations and actions related to law enforcement or criminal incidents

No Comments

No Known Activations