INDEX

Explanations

mentions of individuals' names or related variations

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

-0.77

witz

-0.75

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.71

 warr

-0.68

hood

-0.67

DragonMagazine

-0.62

 Taken

-0.62

hole

-0.61

¶ħ

-0.59

sburg

-0.58

POSITIVE LOGITS

otine

1.18

chio

1.18

ulously

1.13

ulous

1.10

entric

1.06

henko

1.06

helle

1.05

hes

1.05

ulate

1.04

orp

1.02

Activations Density 0.837%

mentions of individuals' names or related variations

No Comments

No Known Activations