INDEX

Explanations

references to international relations and political figures

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

âĢ¦âĢ¦

-0.81

âĢ¦..

-0.79

!!!!!!!!

-0.77

 Assignment

-0.77

 fucking

-0.77

Liter

-0.76

omorph

-0.76

Zip

-0.76

âĢ¦]

-0.75

 ðŁĻĤ

-0.75

POSITIVE LOGITS

 controvers

1.16

 blamed

1.12

 contentious

1.06

 accuse

1.03

 critics

1.02

 dogged

1.02

 lawmakers

1.02

 hotly

1.01

 accusations

1.00

Critics

0.99

Activations Density 6.097%

references to international relations and political figures

No Comments

No Known Activations