INDEX

Explanations

references to specific lines or numbered points in a text

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

CVE

-1.02

ndum

-0.70

sburg

-0.70

 salts

-0.70

ĵĺ

-0.69

sylvania

-0.66

ILD

-0.66

ilton

-0.66

âĺħâĺħ

-0.64

CLASSIFIED

-0.64

POSITIVE LOGITS

backer

1.25

ups

1.14

aments

1.11

ages

1.00

up

0.90

breaks

0.86

breaking

0.80

breakers

0.79

draw

0.75

points

0.75

Activations Density 0.633%

references to specific lines or numbered points in a text

No Comments

No Known Activations