INDEX

Explanations

references to famous landmarks, specifically the Eiffel Tower

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_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

anu

-1.09

aml

-1.01

udeb

-0.97

mble

-0.95

vati

-0.95

hran

-0.93

pport

-0.93

asu

-0.92

icz

-0.92

emo

-0.92

POSITIVE LOGITS

 flush

0.70

 Genius

0.70

 Cruise

0.69

 Bullets

0.67

 bailout

0.66

 Islands

0.65

 Playoff

0.63

 Refuge

0.61

 Valley

0.61

 heartbeat

0.61

Activations Density 0.248%

references to famous landmarks, specifically the Eiffel Tower

No Comments

No Known Activations