INDEX

Explanations

specific identifiers, particularly those related to content or lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

antal

-0.16

ureka

-0.15

anggal

-0.14

.CompareTag

-0.14

Kos

-0.14

UGHT

-0.14

oders

-0.14

ampus

-0.14

aina

-0.14

oppel

-0.14

POSITIVE LOGITS

ident

0.15

ÑģÑĸÐ¼

0.15

 heat

0.14

sdk

0.13

 Pink

0.13

Äįin

0.13

-ie

0.13

_DF

0.13

iqueta

0.13

cribe

0.13

Activations Density 0.030%

specific identifiers, particularly those related to content or lists

No Comments

No Known Activations