INDEX

Explanations

references to traditional concepts or practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

arel

-0.15

ings

-0.14

 originals

-0.14

 ÙĪØ§ÙĤØ¹

-0.14

bras

-0.14

 particular

-0.14

ropolis

-0.13

fal

-0.13

ÑĢÐ°

-0.13

original

-0.13

POSITIVE LOGITS

ists

0.38

ist

0.32

ism

0.24

ISTS

0.24

ised

0.23

ista

0.22

isti

0.22

istic

0.22

izing

0.22

ising

0.22

Activations Density 0.033%

references to traditional concepts or practices

No Comments

No Known Activations