INDEX

Explanations

exclamatory statements or expressions

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

undai

-0.84

 destro

-0.72

 conduc

-0.70

 necessities

-0.70

 utilitarian

-0.69

 glim

-0.67

yrus

-0.65

 restraints

-0.65

matically

-0.65

 plur

-0.65

POSITIVE LOGITS

@#&

1.56

#$

1.30

?!

1.21

:)

1.15

@#

1.13

;)

1.06

 ðŁĺ

1.06

ãĢį

1.05

 ðŁĻĤ

1.04

[/

1.03

Activations Density 0.791%

exclamatory statements or expressions

No Comments

No Known Activations