INDEX

Explanations

actions and activities related to personal engagements and endeavors

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

cia

-0.69

bal

-0.69

eka

-0.68

croft

-0.68

aun

-0.65

lat

-0.63

sac

-0.62

ritical

-0.62

peria

-0.62

phies

-0.60

POSITIVE LOGITS

 furiously

0.76

 lately

0.72

 frantically

0.71

 dangerously

0.70

redients

0.70

 pains

0.68

lass

0.66

GGGGGGGG

0.66

 awfully

0.64

 ducks

0.64

Activations Density 1.424%

actions and activities related to personal engagements and endeavors

No Comments

No Known Activations