INDEX

Explanations

numerical values, particularly related to measurements or specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 Kraj

-0.15

otel

-0.14

ooke

-0.14

eden

-0.14

ãĤ¯ãĥĪ

-0.14

olid

-0.14

UTERS

-0.13

.Apis

-0.13

 Seks

-0.13

cud

-0.13

POSITIVE LOGITS

Ð²Ð¸Ñī

0.15

px

0.14

rig

0.14

Rig

0.14

ippi

0.14

Bij

0.14

ethnic

0.14

 Nisan

0.14

allen

0.14

imo

0.13

Activations Density 0.247%

numerical values, particularly related to measurements or specifications

No Comments

No Known Activations