INDEX

Explanations

instructions or steps related to positioning or future actions

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

Pwr

-0.80

PORT

-0.68

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

-0.68

Merit

-0.67

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.64

advertising

-0.61

HY

-0.61

KN

-0.60

cli

-0.59

MAG

-0.59

POSITIVE LOGITS

ally

1.15

etheless

0.96

 generations

0.94

iations

0.94

 iterations

0.90

ality

0.87

 stages

0.85

iation

0.84

 phases

0.84

noon

0.81

Activations Density 0.030%

instructions or steps related to positioning or future actions

No Comments

No Known Activations

instructions or steps related to positioning or future actions

No Comments

No Known Activations