INDEX

Explanations

terms related to heat

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

ystem

-0.78

oru

-0.74

rahim

-0.74

eca

-0.71

RB

-0.71

Transaction

-0.71

 guiActiveUn

-0.70

utan

-0.70

vous

-0.69

ledged

-0.69

POSITIVE LOGITS

 temperatures

1.18

 hotter

1.01

 oven

1.00

hens

0.97

 climates

0.96

achine

0.94

 humid

0.94

 temperature

0.92

 baths

0.90

 stove

0.89

Activations Density 0.942%

terms related to heat

No Comments

No Known Activations