INDEX

Explanations

proper nouns related to locations and names

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_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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 reckoning

-0.58

 unsur

-0.57

 tremend

-0.57

 scanner

-0.56

 echo

-0.56

 Kenobi

-0.56

 Hate

-0.55

FML

-0.55

 recoil

-0.54

XM

-0.54

POSITIVE LOGITS

cled

0.88

pillar

0.88

berus

0.81

estine

0.80

ading

0.80

tesy

0.80

cling

0.79

rier

0.78

cil

0.77

perate

0.75

Activations Density 0.054%

proper nouns related to locations and names

No Comments

No Known Activations