INDEX

Explanations

the word "were" in various contexts and forms

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

ston

-0.16

 Ø¬Ø§Ùħ

-0.15

ë

-0.14

-syntax

-0.14

kin

-0.14

EAR

-0.14

 disfr

-0.14

ERY

-0.14

Jar

-0.13

yar

-0.13

POSITIVE LOGITS

ìľ¨

0.16

 Leban

0.16

obar

0.15

elize

0.15

áº¥p

0.15

uant

0.15

ileaks

0.15

ÑĩÐ¸Ñģ

0.14

ivec

0.14

_DBG

0.14

Activations Density 0.085%

the word "were" in various contexts and forms

No Comments

No Known Activations