INDEX

Explanations

unique identifiers related to various subjects or contexts, potentially focusing on specifics in datasets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

tul

-0.14

iel

-0.14

 Jones

-0.14

 praw

-0.14

.inputs

-0.14

_navigation

-0.13

[âĢ¦

-0.13

UGHT

-0.13

oby

-0.13

ForObject

-0.13

POSITIVE LOGITS

autos

0.15

stad

0.15

[from

0.15

cz

0.14

bos

0.14

ÑĭÑĪ

0.14

hower

0.14

ÑģÑĥÑĤ

0.14

zek

0.14

.mit

0.13

Activations Density 0.018%

unique identifiers related to various subjects or contexts, potentially focusing on specifics in datasets

No Comments

No Known Activations