INDEX

Explanations

terms indicating established knowledge or well-documented information

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

gro

-0.16

raud

-0.16

aba

-0.15

erton

-0.15

Ð½Ð¸Ñĩ

-0.15

Gro

-0.14

 alle

-0.14

ouch

-0.14

aldi

-0.14

WARDED

-0.14

POSITIVE LOGITS

TRL

0.16

jit

0.15

-Clause

0.15

à¥įà¤¸à¤°

0.14

ÑģÑĮ

0.14

 dign

0.14

LineStyle

0.14

ocol

0.14

rops

0.14

ãĥ³ãĤº

0.14

Activations Density 0.039%

terms indicating established knowledge or well-documented information

No Comments

No Known Activations