INDEX

Explanations

dialogue tags of people speaking

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

cies

-0.71

rib

-0.68

hiba

-0.65

emo

-0.65

hid

-0.64

ingen

-0.63

wire

-0.63

Compan

-0.62

ancies

-0.61

enne

-0.61

POSITIVE LOGITS

 âĢ

1.22

++

1.15

 Â±

1.14

***

1.13

[*]

1.11

 =====

1.11

 âĹı

1.08

<+

1.08

1.07

Activations Density 1.105%

dialogue tags of people speaking

No Comments

No Known Activations