INDEX

Explanations

phrases indicating the concept of reasonableness or fairness

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

icket

-0.73

chet

-0.70

chu

-0.70

planes

-0.69

yi

-0.69

frey

-0.69

berries

-0.68

stals

-0.65

cart

-0.64

flower

-0.63

POSITIVE LOGITS

tarian

0.98

acies

0.82

 inference

0.76

 excuse

0.74

 precaution

0.72

 expectation

0.71

soType

0.70

ufact

0.70

 justification

0.70

itably

0.70

Activations Density 0.959%

phrases indicating the concept of reasonableness or fairness

No Comments

No Known Activations