INDEX

Explanations

mentions of legal and justice-related entities and topics

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

iple

-0.79

irst

-0.74

igslist

-0.72

hetically

-0.70

acid

-0.70

uring

-0.69

uers

-0.68

ulators

-0.68

hing

-0.67

ulates

-0.67

POSITIVE LOGITS

 Scalia

0.89

 GOODMAN

0.75

 Dept

0.75

justice

0.74

 Gorsuch

0.72

 Sonia

0.71

 Department

0.71

 Brennan

0.71

Justice

0.71

iciary

0.71

Activations Density 0.628%

mentions of legal and justice-related entities and topics

No Comments

No Known Activations