INDEX

Explanations

instances of news articles or reports describing criminal activities and events

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

avorite

-0.71

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.70

 stockp

-0.69

zees

-0.69

pav

-0.67

 Malfoy

-0.66

sold

-0.65

atorium

-0.64

inventory

-0.64

itures

-0.64

POSITIVE LOGITS

News

1.22

 News

1.19

 Daily

1.18

 reporters

1.16

Journal

1.14

 Reporter

1.13

Daily

1.09

aretz

1.09

 Tribune

1.05

 Investigative

1.04

Activations Density 1.600%

instances of news articles or reports describing criminal activities and events

No Comments

No Known Activations