INDEX

Explanations

phrases related to data analysis and research methods

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

mare

-0.84

mods

-0.80

 Rebell

-0.76

itol

-0.75

 Shutdown

-0.74

 Revival

-0.73

SourceFile

-0.73

ardless

-0.71

 Happy

-0.71

Owner

-0.71

POSITIVE LOGITS

 analys

1.24

 glean

1.22

 insights

1.21

 decipher

1.19

 unbiased

1.19

 analytical

1.19

 documenting

1.17

 analyzing

1.14

 analyses

1.13

 analysis

1.10

Activations Density 3.333%

phrases related to data analysis and research methods

No Comments

No Known Activations