INDEX

Explanations

phrases related to direction or movement

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

ollar

-0.62

manship

-0.60

pots

-0.60

 lett

-0.58

iqueness

-0.58

ãĤ¦ãĤ¹

-0.57

pot

-0.56

 situational

-0.56

Ability

-0.56

 Actual

-0.56

POSITIVE LOGITS

 towards

0.99

 toward

0.98

 stairs

0.98

 downhill

0.95

Ô

0.94

 unnoticed

0.94

 corridors

0.88

wards

0.86

 blindly

0.85

 unch

0.85

Activations Density 2.739%

phrases related to direction or movement

No Comments

No Known Activations