INDEX

Explanations

information related to post-event occurrences or outcomes

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ¤ãĥĪ

-0.80

aez

-0.75

isible

-0.74

tick

-0.71

Especially

-0.71

Enabled

-0.70

alf

-0.70

Scale

-0.70

Tur

-0.69

ãĥ³ãĤ¸

-0.68

POSITIVE LOGITS

 completing

1.04

 consulting

0.94

math

0.94

market

0.92

 receiving

0.91

 defeating

0.88

 consultation

0.88

 obtaining

0.86

 noticing

0.83

thought

0.82

Activations Density 0.110%

information related to post-event occurrences or outcomes

No Comments

No Known Activations