INDEX

Explanations

locations or political entities

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

BOOK

-0.76

ynt

-0.75

fen

-0.74

byn

-0.70

colo

-0.70

htar

-0.70

henko

-0.67

chel

-0.67

 Britann

-0.66

deck

-0.65

POSITIVE LOGITS

 State

1.14

 Springs

1.09

Gov

1.08

 Governor

1.06

sylvania

0.98

 Attorney

0.93

 Legislature

0.93

 Tech

0.91

 Rapids

0.91

 Avenue

0.89

Activations Density 0.659%

locations or political entities

No Comments

No Known Activations