INDEX

Explanations

occurrences of the word "en-" and its variations, indicating a focus on words related to involvement or participation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ÙħØ§ÙĨÛĮ

-0.18

intestinal

-0.16

cular

-0.16

phan

-0.16

Ø¶ÙĬ

-0.15

weise

-0.15

gende

-0.15

guns

-0.15

fak

-0.15

ulence

-0.15

POSITIVE LOGITS

esa

0.16

uru

0.16

wine

0.15

Ð½Ð°ÑĩÐµ

0.15

CRYPT

0.15

eed

0.15

Åĳs

0.15

prises

0.15

ÏĥÎ¿Ïħ

0.15

OUGH

0.14

Activations Density 0.066%

occurrences of the word "en-" and its variations, indicating a focus on words related to involvement or participation

No Comments

No Known Activations