INDEX

Explanations

possessive pronouns or forms indicating ownership

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

unin

-0.85

$$$$

-0.81

ulkan

-0.78

Widget

-0.75

ãħĭ

-0.74

amiya

-0.74

\\\\

-0.74

naires

-0.73

holes

-0.73

Frameworks

-0.73

POSITIVE LOGITS

 Jonah

0.98

 Investigative

0.91

 Juliet

0.89

 correspondent

0.88

 Matthew

0.87

own

0.87

 Belief

0.86

 Adam

0.86

 editor

0.85

 Andrew

0.84

Activations Density 0.028%

possessive pronouns or forms indicating ownership

No Comments

No Known Activations