INDEX

Explanations

surprising elements or events

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

burgh

-0.75

claimer

-0.73

rio

-0.73

eworks

-0.72

©¶æ

-0.69

onest

-0.68

apers

-0.66

otent

-0.66

onse

-0.66

tein

-0.66

POSITIVE LOGITS

 coincidence

1.01

ly

0.92

 twists

0.87

 occurrences

0.87

 occurrence

0.83

 coinc

0.82

 phenomenon

0.80

 phenomena

0.79

 Flavoring

0.78

 juxtap

0.77

Activations Density 2.268%

surprising elements or events

No Comments

No Known Activations