INDEX

Explanations

descriptive words related to enhancing or diversifying a specific item or situation

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

akedown

-0.74

onga

-0.66

 Burk

-0.64

 Sect

-0.61

jay

-0.60

 Lich

-0.60

 Bans

-0.59

asel

-0.59

 majority

-0.57

ç«

-0.57

POSITIVE LOGITS

fulness

1.15

iness

1.13

lessness

0.99

ability

0.93

ier

0.91

abl

0.90

iveness

0.89

smanship

0.89

uality

0.86

abilities

0.84

Activations Density 0.251%

descriptive words related to enhancing or diversifying a specific item or situation

No Comments

No Known Activations