INDEX

Explanations

the term "simple" and related simple or basic concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

hawk

-0.16

oller

-0.16

ipur

-0.16

ntax

-0.15

urement

-0.15

mey

-0.15

orden

-0.14

ogi

-0.14

utto

-0.14

znam

-0.14

POSITIVE LOGITS

 Jeremy

0.16

lak

0.16

ichel

0.15

 reconcile

0.14

 reconc

0.14

 icon

0.14

uese

0.14

Ð¾ÑģÑĤ

0.14

 rooms

0.14

Interpolator

0.14

Activations Density 0.023%

the term "simple" and related simple or basic concepts

No Comments

No Known Activations