INDEX

Explanations

descriptions involving personal experiences and reflections

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

evidence

-0.75

Events

-0.73

Contents

-0.71

reports

-0.71

overs

-0.70

Edit

-0.67

iments

-0.66

items

-0.65

ATURES

-0.65

grounds

-0.65

POSITIVE LOGITS

 bunch

1.27

lot

1.25

 handful

1.10

rouse

1.10

uras

1.07

 couple

1.07

 multitude

1.06

few

1.04

 single

1.01

 piece

0.99

Activations Density 1.750%

descriptions involving personal experiences and reflections

No Comments

No Known Activations