INDEX

Explanations

references to nature and outdoor activities

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

oned

-0.75

onest

-0.75

oning

-0.74

aged

-0.73

rano

-0.71

rations

-0.68

itone

-0.68

rican

-0.67

acters

-0.66

enger

-0.66

POSITIVE LOGITS

 preserves

0.90

 Conserv

0.83

 sciences

0.79

 conservation

0.79

 instinct

0.74

folk

0.74

 fiction

0.74

 habitat

0.73

 vegetation

0.73

 flora

0.73

Activations Density 0.016%

references to nature and outdoor activities

No Comments

No Known Activations