INDEX

Explanations

URLs and links within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 narrowly

-0.15

iona

-0.14

_serializer

-0.14

bett

-0.14

adal

-0.14

 Pick

-0.13

 longevity

-0.13

pick

-0.13

 entirely

-0.13

ahoma

-0.13

POSITIVE LOGITS

 Binder

0.16

entes

0.16

ITIZE

0.15

 ÑĢÐ°Ð¹

0.15

anders

0.15

HEME

0.15

asses

0.15

TEGER

0.15

Bind

0.14

iddi

0.14

Activations Density 0.004%

URLs and links within the text

No Comments

No Known Activations