INDEX

Explanations

phrases related to personal emotional experiences and reflections

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

arthed

-0.83

 Policies

-0.78

specified

-0.77

 patents

-0.77

idelines

-0.77

senal

-0.77

 Manufacturing

-0.76

available

-0.74

winner

-0.74

ospons

-0.73

POSITIVE LOGITS

 sadness

1.47

 numb

1.39

 feeling

1.35

 sorrow

1.35

 tears

1.29

 longing

1.28

 anguish

1.28

 agony

1.26

 remembering

1.25

 waking

1.24

Activations Density 5.274%

phrases related to personal emotional experiences and reflections

No Comments

No Known Activations