INDEX

Explanations

keywords related to significant topics or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

ehr

-0.15

 pari

-0.15

eprom

-0.14

oque

-0.14

brate

-0.13

ixa

-0.13

å£«

-0.13

Benchmark

-0.13

.getElements

-0.13

ÎºÏģÎ±

-0.13

POSITIVE LOGITS

antz

0.15

 importance

0.15

 important

0.15

rou

0.15

hait

0.14

eneg

0.14

 Lessons

0.14

macro

0.14

_HELPER

0.14

 single

0.14

Activations Density 0.037%

keywords related to significant topics or concepts

No Comments

No Known Activations