INDEX

Explanations

adjectives describing negative or harsh characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aton

-0.15

avan

-0.14

roy

-0.14

zig

-0.14

wn

-0.13

_multiplier

-0.13

 trimest

-0.13

berger

-0.13

ern

-0.13

POSITIVE LOGITS

ADDE

0.15

@student

0.14

unsch

0.13

Fmt

0.13

/***************************************************************************↵

0.13

etc

0.13

-gnu

0.13

eyse

0.13

ÏĥÏĦÎ¬

0.13

CHANT

0.13

Activations Density 0.085%

adjectives describing negative or harsh characteristics

No Comments

No Known Activations