INDEX

Explanations

phrases related to political and social issues

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

ITNESS

-0.67

interstitial

-0.66

 consolation

-0.64

VIDIA

-0.60

ilant

-0.59

annot

-0.59

 Athe

-0.58

grand

-0.58

omorphic

-0.58

 congr

-0.58

POSITIVE LOGITS

 altogether

1.12

 prematurely

0.89

 accordingly

0.87

 indefinitely

0.87

 unnecessarily

0.85

 ASAP

0.81

 lest

0.78

 amid

0.77

 considerably

0.75

 into

0.73

Activations Density 2.782%

phrases related to political and social issues

No Comments

No Known Activations