INDEX

Explanations

expressions of disagreement

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

 protected

-0.67

 elimination

-0.63

 learning

-0.62

 Classification

-0.60

 Butt

-0.59

 scrambling

-0.59

 circ

-0.58

 guided

-0.58

 Carth

-0.58

 ejected

-0.58

POSITIVE LOGITS

't

1.68

ÃŃ

1.02

Â´

0.98

ned

0.97

ates

0.96

etsk

0.94

uts

0.93

hips

0.93

ate

0.91

eness

0.91

Activations Density 0.118%

expressions of disagreement

No Comments

No Known Activations