INDEX

Explanations

discussions about depression and mental health conditions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åĪĳ

-0.18

amage

-0.15

inj

-0.15

 injuries

-0.14

ALER

-0.14

juries

-0.14

Ã®tre

-0.14

é¾Ħ

-0.14

tica

-0.14

 ÑĤÑĢÑĥ

-0.14

POSITIVE LOGITS

 depression

0.54

 depressive

0.46

 Depression

0.44

 depressed

0.44

Dep

0.44

 depress

0.43

Dep

0.41

 antidepress

0.37

 mood

0.35

dep

0.35

Activations Density 0.091%

discussions about depression and mental health conditions

No Comments

No Known Activations