INDEX

Explanations

descriptions of outcomes or results

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

osponsors

-0.67

udi

-0.65

 Passage

-0.65

 blat

-0.64

don

-0.64

 actionGroup

-0.63

NetMessage

-0.62

conservancy

-0.62

afort

-0.62

hare

-0.61

POSITIVE LOGITS

iveness

0.97

 thereof

0.94

ivity

0.85

ively

0.81

 result

0.75

ivism

0.73

of

0.72

ivist

0.72

aries

0.71

arian

0.69

Activations Density 0.042%

descriptions of outcomes or results

No Comments

No Known Activations