INDEX

Explanations

phrases related to individuals, particularly focusing on the name "Jean"

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

holders

-0.77

itational

-0.76

ulnerability

-0.74

ield

-0.65

SUP

-0.65

 Territories

-0.64

ulnerable

-0.62

tsky

-0.62

houses

-0.61

kish

-0.60

POSITIVE LOGITS

ette

0.83

-+

0.81

erate

0.78

regate

0.78

anson

0.78

ements

0.76

otti

0.74

stadt

0.74

unciation

0.72

rador

0.72

Activations Density 0.502%

phrases related to individuals, particularly focusing on the name "Jean"

No Comments

No Known Activations