INDEX

Explanations

terms related to perception in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

mers

-0.73

eni

-0.70

chn

-0.66

enegger

-0.66

loads

-0.65

stead

-0.65

sch

-0.65

drivers

-0.64

ilant

-0.62

 Sieg

-0.61

POSITIVE LOGITS

 perceptions

0.87

 perception

0.86

 Perception

0.79

 biases

0.79

 bias

0.76

wcsstore

0.75

ally

0.73

 disson

0.72

 impression

0.72

 impressions

0.70

Activations Density 0.011%

terms related to perception in various contexts

No Comments

No Known Activations