INDEX

Explanations

words related to locations, people, or significant entities in a narrative context

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

iline

-0.15

orny

-0.15

ÑĢÐ°Ðº

-0.15

ervas

-0.15

rett

-0.15

 Drew

-0.14

rend

-0.14

ç·ł

-0.14

Ð»Ð°ÑģÑĤÐ¸

-0.14

allah

-0.14

POSITIVE LOGITS

ÑĩÐ¸

0.14

apis

0.14

kum

0.14

 workings

0.13

 starred

0.13

 reversal

0.13

bul

0.13

_INTR

0.13

bÄĽ

0.13

Activations Density 0.031%

words related to locations, people, or significant entities in a narrative context

No Comments

No Known Activations