INDEX

Explanations

names of locations, especially cities or regions

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

ADRA

-0.68

catentry

-0.66

 tremend

-0.66

 pregn

-0.65

 contrace

-0.61

 defect

-0.60

 Sabha

-0.60

 illum

-0.60

ãĤ©

-0.60

 bluff

-0.59

POSITIVE LOGITS

velt

0.83

lein

0.76

eworthy

0.76

rences

0.74

emp

0.72

hent

0.72

anamo

0.72

enges

0.72

phant

0.71

ghan

0.71

Activations Density 1.619%

names of locations, especially cities or regions

No Comments

No Known Activations