INDEX

Explanations

Logitech

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

terday

-0.90

xual

-0.69

 Kimmel

-0.59

ETHOD

-0.58

ãĥĨãĤ£

-0.57

HAEL

-0.56

riages

-0.56

 tort

-0.56

 cour

-0.56

 injuring

-0.55

POSITIVE LOGITS

ging

1.45

gers

1.44

istics

1.32

istical

1.25

ged

1.21

itech

1.18

ger

1.16

istically

1.15

icians

1.05

istic

1.04

Activations Density 0.013%

Logitech

No Comments

No Known Activations