INDEX

Explanations

instances of personal actions and consequences within a narrative context

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

 Manifest

-0.76

 Liqu

-0.62

 cultiv

-0.62

 Commun

-0.59

 Communities

-0.59

 Written

-0.58

 Measures

-0.58

 Implementation

-0.58

 Capital

-0.57

 intimid

-0.57

POSITIVE LOGITS

now

1.06

 hindsight

0.91

now

0.91

 regretted

0.80

NOW

0.79

 remembered

0.77

enegger

0.75

mud

0.74

cedented

0.74

Fast

0.74

Activations Density 0.708%

instances of personal actions and consequences within a narrative context

No Comments

No Known Activations