INDEX

Explanations

occurrences of the word "in" along with related context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

avian

-0.18

 TMPro

-0.15

æ¹

-0.15

èĭĹ

-0.15

ality

-0.14

 interven

-0.14

oad

-0.14

 btnSave

-0.14

aton

-0.13

wen

-0.13

POSITIVE LOGITS

iversit

0.16

anium

0.16

ioxide

0.15

ÑĥÑĢÐ°

0.15

ctors

0.15

ouro

0.15

ç¾©

0.15

regon

0.15

esis

0.14

INAL

0.14

Activations Density 0.012%

occurrences of the word "in" along with related context

No Comments

No Known Activations