INDEX

Explanations

technical language or analysis related to mechanical systems and experimental setup

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

 enjoyment

-0.66

BuyableInstoreAndOnline

-0.64

 healed

-0.59

 condol

-0.59

 autism

-0.59

 Leilan

-0.58

stable

-0.58

laugh

-0.58

 Happiness

-0.57

 diaper

-0.57

POSITIVE LOGITS

requires

0.97

apo

0.86

 requires

0.85

 require

0.81

must

0.74

 resort

0.73

require

0.73

 must

0.72

 recourse

0.72

OSE

0.71

Activations Density 1.414%

technical language or analysis related to mechanical systems and experimental setup

No Comments

No Known Activations