INDEX

Explanations

references to auditory impairments

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

xon

-0.70

 Values

-0.68

 Secret

-0.67

 Bundy

-0.66

 Faction

-0.66

ulkan

-0.65

 guiActiveUnfocused

-0.65

anova

-0.65

AU

-0.64

 Andromeda

-0.62

POSITIVE LOGITS

ening

1.21

ness

1.13

ened

1.06

 deaf

1.02

 ears

0.96

blind

0.88

nesses

0.87

phas

0.86

eners

0.84

bread

0.84

Activations Density 0.022%

references to auditory impairments

No Comments

No Known Activations