INDEX

Explanations

mentions or references encouraging someone to remember or be mindful of certain information or tips

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ode

-0.80

=>

-0.67

NESS

-0.66

assi

-0.66

atana

-0.66

iliary

-0.66

ooter

-0.66

KEN

-0.66

KY

-0.66

acus

-0.66

POSITIVE LOGITS

 tabs

1.24

 pace

1.11

 afloat

1.08

 track

1.02

 secrets

0.93

 vigil

0.88

 alive

0.88

 quiet

0.82

 kosher

0.78

 secret

0.78

Activations Density 0.955%

mentions or references encouraging someone to remember or be mindful of certain information or tips

No Comments

No Known Activations