INDEX

Explanations

specific names or entities, such as people or organizations

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

SpaceEngineers

-0.75

 newsp

-0.69

udeb

-0.68

é¾

-0.67

Magikarp

-0.66

catentry

-0.66

Ó

-0.65

ebted

-0.65

Reviewer

-0.62

ãģ¾

-0.62

POSITIVE LOGITS

hower

0.88

 Profile

0.81

itably

0.71

osate

0.71

wagen

0.66

mania

0.65

 Grove

0.65

 Friendly

0.64

fly

0.64

croft

0.64

Activations Density 1.328%

specific names or entities, such as people or organizations

No Comments

No Known Activations