INDEX

Explanations

negative words, particularly adjectives describing things as stupid

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-2.72

/***

-0.73

ⓧ

-0.73

-0.72

HasIndex

-0.70

})();

-0.70

/*!

-0.63

define

-0.63

 насељу

-0.62

///**

-0.61

POSITIVE LOGITS

 tramont

1.55

 sappi

1.39

 stockholm

1.25

 maroc

1.22

 milano

1.17

 ibiza

1.16

 riviera

1.15

 Khart

1.15

 cristina

1.14

 lidl

1.13

Activations Density 0.156%

negative words, particularly adjectives describing things as stupid

No Comments

No Known Activations