INDEX

Explanations

terms related to economic or social status, particularly highlighting the concept of poverty

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

ategory

-1.00

agall

-0.78

thus

-0.75

CU

-0.75

natureconservancy

-0.74

CRE

-0.70

IDA

-0.70

auer

-0.69

 enthusi

-0.68

FY

-0.68

POSITIVE LOGITS

 peasants

0.89

lings

0.87

 souls

0.83

 performers

0.78

 peasant

0.76

 subsistence

0.75

folk

0.72

 poorest

0.71

child

0.71

liction

0.71

Activations Density 0.242%

terms related to economic or social status, particularly highlighting the concept of poverty

No Comments

No Known Activations