INDEX

Explanations

specific phrases or words related to being correct, effective, or appropriate

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

ĸļ

-0.83

ADRA

-0.72

anned

-0.71

cit

-0.70

ruary

-0.69

ushima

-0.67

ivism

-0.65

bery

-0.65

 Lilly

-0.64

oute

-0.63

POSITIVE LOGITS

 amount

1.11

 combination

0.92

 balance

0.87

 thing

0.84

 sized

0.83

 antidote

0.82

 attitude

0.82

 kind

0.81

 circumstances

0.81

 temperament

0.81

Activations Density 0.506%

specific phrases or words related to being correct, effective, or appropriate

No Comments

No Known Activations