INDEX

Explanations

phrases related to personal stories or experiences, especially those involving personal conflict or struggles

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

 Travels

-0.65

aw

-0.64

 activ

-0.61

eele

-0.59

 exerc

-0.55

ize

-0.55

 slee

-0.54

 advoc

-0.54

 conven

-0.54

 culminated

-0.54

POSITIVE LOGITS

nor

1.91

Nor

1.59

nor

1.57

 Instead

1.36

Nor

1.36

Instead

1.27

yet

1.26

 Neither

1.20

 anymore

1.18

unless

1.13

Activations Density 2.398%

phrases related to personal stories or experiences, especially those involving personal conflict or struggles

No Comments

No Known Activations