INDEX

Explanations

statements related to criticism or analysis of official reports or documents

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

Enlarge

-0.69

window

-0.68

folk

-0.66

angular

-0.66

ãĥİ

-0.64

brance

-0.63

Copy

-0.62

mt

-0.62

etric

-0.61

immune

-0.60

POSITIVE LOGITS

 sarcast

1.11

 bluntly

1.02

 emphatically

0.95

 rhet

0.93

 omin

0.90

 noting

0.89

:"

0.88

 passionately

0.87

 confidently

0.85

 secondly

0.83

Activations Density 1.014%

statements related to criticism or analysis of official reports or documents

No Comments

No Known Activations