INDEX

Explanations

legal contexts and discussions about reasons or justifications

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

semble

-0.74

 Carbuncle

-0.73

chron

-0.65

KY

-0.63

agog

-0.59

lasses

-0.58

ymph

-0.58

ibaba

-0.58

omez

-0.57

 Lans

-0.57

POSITIVE LOGITS

why

1.38

abl

1.21

why

1.11

WHY

1.10

 behind

0.92

 justifying

0.91

 rationale

0.85

Why

0.81

 cele

0.81

Why

0.81

Activations Density 0.502%

legal contexts and discussions about reasons or justifications

No Comments

No Known Activations