INDEX

Explanations

code-related terms and software instructions

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

eteria

-0.76

 Casey

-0.76

 Manson

-0.74

FML

-0.71

 Pearce

-0.69

 screenings

-0.68

 Sachs

-0.68

 Mercy

-0.68

 reception

-0.68

JPM

-0.67

POSITIVE LOGITS

chance

1.18

ebook

1.13

tro

1.11

dict

1.09

blank

1.07

unique

1.05

scene

1.04

enabled

1.03

relevant

1.03

tra

1.03

Activations Density 0.204%

code-related terms and software instructions

No Comments

No Known Activations