INDEX

Explanations

references to specific names or titles

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lessly

-0.87

voc

-0.83

 tremend

-0.81

 includ

-0.79

aciously

-0.75

antically

-0.73

lessness

-0.71

resent

-0.71

ickr

-0.70

atically

-0.69

POSITIVE LOGITS

 Ones

1.20

 Feet

1.11

 Forces

1.10

 Mile

1.09

 Lives

1.06

 Moves

1.01

 Girl

1.01

 Roads

1.01

 Skies

1.01

 Points

1.01

Activations Density 1.299%

references to specific names or titles

No Comments

No Known Activations