INDEX

Explanations

terms related to imagination and its various forms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

bye

-0.70

andra

-0.69

ldon

-0.67

upon

-0.67

[+

-0.67

paying

-0.67

hill

-0.66

hide

-0.66

×Ĳ

-0.66

ishops

-0.65

POSITIVE LOGITS

 imagination

1.05

 imag

1.00

 imagin

0.88

issance

0.88

 imagining

0.82

 Balloon

0.75

 Interpret

0.74

 imaginative

0.73

ufact

0.71

urable

0.71

Activations Density 0.040%

terms related to imagination and its various forms

No Comments

No Known Activations