INDEX

Explanations

words related to negative criticism or controversy

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

 Skydragon

-0.81

selves

-0.81

RAL

-0.68

HCR

-0.62

*/(

-0.62

ansas

-0.59

 Built

-0.59

 disclaim

-0.59

osphere

-0.59

EngineDebug

-0.58

POSITIVE LOGITS

orescence

1.15

orescent

1.08

oresc

0.99

ickr

0.98

oyd

0.94

uffy

0.89

atable

0.86

icker

0.85

iers

0.84

keyes

0.83

Activations Density 2.345%

words related to negative criticism or controversy

No Comments

No Known Activations