INDEX

Explanations

the verb "have" in various contexts and forms

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

etting

-0.69

oshi

-0.62

osi

-0.60

ensing

-0.58

etter

-0.56

andan

-0.56

alias

-0.54

note

-0.54

TG

-0.53

 Quote

-0.52

POSITIVE LOGITS

 been

1.26

 gotten

1.11

been

1.10

 undergone

1.04

 recourse

0.98

 Been

0.96

 gone

0.95

gotten

0.94

stood

0.93

 begun

0.90

Activations Density 1.074%

the verb "have" in various contexts and forms

No Comments

No Known Activations