INDEX

Explanations

URLs and web links in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

endra

-0.17

pler

-0.15

itness

-0.14

undra

-0.14

sein

-0.14

 Bers

-0.14

ogg

-0.14

APTER

-0.14

ype

-0.14

eria

-0.13

POSITIVE LOGITS

doi

0.30

dx

0.25

DOI

0.23

doi

0.22

resolver

0.22

dx

0.21

uir

0.20

hdl

0.18

ContentView

0.18

DOI

0.17

Activations Density 0.012%

URLs and web links in the text

No Comments

No Known Activations