INDEX

Explanations

references to individuals' names and affiliations in a specific context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

cus

-0.15

Ju

-0.15

okie

-0.15

ursions

-0.14

lanma

-0.14

opia

-0.14

 Loud

-0.14

Ã¡p

-0.13

 ballistic

-0.13

Ju

-0.13

POSITIVE LOGITS

brero

0.15

opro

0.15

:<?

0.15

oleon

0.14

mare

0.14

elen

0.14

æŁ±

0.14

krv

0.14

_GRE

0.14

esson

0.14

Activations Density 0.105%

references to individuals' names and affiliations in a specific context

No Comments

No Known Activations