INDEX

Explanations

phrases related to political figures and policies

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

 mats

-0.75

 convent

-0.72

ENE

-0.71

AGES

-0.70

 Seraph

-0.70

Ort

-0.68

 Sisters

-0.68

 Shib

-0.66

 Collective

-0.66

 Sensor

-0.65

POSITIVE LOGITS

 veto

1.13

 appoint

1.13

 pard

1.11

 pardon

1.05

 vetoed

1.03

 campaigned

0.98

 appointing

0.96

 aides

0.96

enegger

0.90

 impeachment

0.88

Activations Density 2.903%

phrases related to political figures and policies

No Comments

No Known Activations