INDEX

Explanations

elements related to user interactions and compliance with guidelines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

-Token

-0.15

âĢª

-0.15

suite

-0.13

_DEPRECATED

-0.13

/Dk

-0.13

ittance

-0.13

raisal

-0.13

$core

-0.13

-Core

-0.12

kola

-0.12

POSITIVE LOGITS

gether

0.27

itionally

0.21

bidden

0.20

nger

0.19

theless

0.19

:///

0.18

stantiate

0.17

Â·

0.16

oretical

0.15

tempts

0.15

Activations Density 1.504%

elements related to user interactions and compliance with guidelines

No Comments

No Known Activations