INDEX

Explanations

technology-related terms or computer programming code snippets

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

olicy

-0.76

æĸ¹

-0.73

 Debor

-0.73

ERC

-0.65

 Radius

-0.63

ItemTracker

-0.62

ELS

-0.62

riad

-0.60

ãĥ¼ãĥĨãĤ£

-0.60

ALK

-0.60

POSITIVE LOGITS

ibrary

1.21

oaded

1.19

phe

1.19

anguage

1.15

ength

1.14

inda

1.10

phia

1.09

uxe

1.08

iberal

1.06

ibr

1.04

Activations Density 1.849%

technology-related terms or computer programming code snippets

No Comments

No Known Activations