INDEX

Explanations

references to relationships and contrasts between inside and outside elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 domestically

-0.19

 inland

-0.17

 underlying

-0.15

 locally

-0.15

.getLocal

-0.14

 Domestic

-0.13

onium

-0.13

 centrally

-0.13

 privately

-0.13

oplan

-0.12

POSITIVE LOGITS

 outside

1.10

å¤ĸ

1.01

 Outside

0.98

outside

0.95

Outside

0.93

 exterior

0.91

 external

0.88

 å¤ĸ

0.81

 outer

0.78

external

0.78

Activations Density 0.333%

references to relationships and contrasts between inside and outside elements

No Comments

No Known Activations

references to relationships and contrasts between inside and outside elements

No Comments

No Known Activations