INDEX

Explanations

words related to long-term considerations or implications

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IRO

-0.80

illard

-0.75

ILLE

-0.71

leck

-0.70

 Compass

-0.69

atche

-0.63

Always

-0.63

externalActionCode

-0.61

ILA

-0.61

ONSORED

-0.60

POSITIVE LOGITS

itud

1.29

itudinal

1.11

lasting

1.10

ago

1.08

sword

1.07

term

1.07

 overdue

1.06

itude

1.05

 term

1.05

 lasting

1.01

Activations Density 1.057%

words related to long-term considerations or implications

No Comments

No Known Activations