INDEX

Explanations

terms related to specific variants and classifications within the context of a health or drug discussion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

okud

-0.18

ongyang

-0.16

imar

-0.16

erable

-0.15

ernen

-0.14

æľ¬

-0.14

Inspectable

-0.14

olla

-0.14

_pet

-0.14

erm

-0.13

POSITIVE LOGITS

ninger

0.15

é¨

0.15

_PKT

0.15

ahun

0.14

 soft

0.14

à¸Ļ

0.14

 Bravo

0.14

Î¯Ïĥ

0.13

Activations Density 0.006%

terms related to specific variants and classifications within the context of a health or drug discussion

No Comments

No Known Activations