INDEX

Explanations

mentions of government officials and bureaucratic terms

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

 Latter

-0.75

Stream

-0.71

 natives

-0.67

 Condition

-0.67

 Contents

-0.64

 Strip

-0.64

 Nation

-0.63

avery

-0.63

 Artists

-0.63

 Methods

-0.62

POSITIVE LOGITS

esses

1.02

ially

1.00

 overseeing

0.93

alty

0.91

hips

0.91

 tasked

0.90

ial

0.90

iate

0.88

ess

0.86

ials

0.86

Activations Density 0.959%

mentions of government officials and bureaucratic terms

No Comments

No Known Activations