INDEX

Explanations

articles of clothing, especially when described in specific situations or symbolic contexts

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

atican

-0.69

ruciating

-0.67

 Galile

-0.67

ithmetic

-0.66

 timelines

-0.65

raltar

-0.65

otten

-0.64

£ı

-0.63

enthal

-0.63

 pregn

-0.62

POSITIVE LOGITS

 worn

1.54

 sleeves

1.31

 wearer

1.31

 sleeve

1.17

 apparel

1.13

 trousers

1.12

 straps

1.11

 shirt

1.11

 jacket

1.10

 shirts

1.08

Activations Density 1.117%

articles of clothing, especially when described in specific situations or symbolic contexts

No Comments

No Known Activations