INDEX

Explanations

references to buildings and structures 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

å§«

-0.70

osate

-0.64

 geopolitical

-0.62

ĺħ

-0.61

¿½

-0.58

 Dele

-0.56

ĪĴ

-0.55

olitics

-0.54

CFR

-0.54

omething

-0.52

POSITIVE LOGITS

 envelope

0.80

 floor

0.77

 perimeter

0.77

 entrance

0.73

 canopy

0.72

atorium

0.71

arium

0.70

's

0.70

 gates

0.70

ifice

0.70

Activations Density 0.276%

references to buildings and structures in various contexts

No Comments

No Known Activations