INDEX

Explanations

pronouns indicating possession

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

vine

-0.77

 Izan

-0.72

ocument

-0.72

haus

-0.67

river

-0.66

 Frazier

-0.65

 Lerner

-0.64

otin

-0.63

bender

-0.62

 largeDownload

-0.62

POSITIVE LOGITS

own

2.07

self

1.25

OWN

1.25

 respective

1.23

selves

1.15

 newfound

1.13

Own

1.12

 rightful

1.11

 usual

1.10

 surroundings

1.03

Activations Density 1.212%

pronouns indicating possession

No Comments

No Known Activations