INDEX

Explanations

phrases related to hiding or concealing something controversial or negative

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>

-2.62

lateinit

-0.79

},{

-0.74

ComponentModel

-0.74

/***

-0.74

HasIndex

-0.70

dst

-0.69

src

-0.68

glTexCoord

-0.67

nmgp

-0.65

POSITIVE LOGITS

 stockholm

1.59

 emphat

1.59

 !...

1.55

 ?...

1.54

vhs

1.54

 milf

1.53

 indestru

1.52

 affor

1.50

 hentai

1.48

 increa

1.47

Activations Density 0.253%

phrases related to hiding or concealing something controversial or negative

No Comments

No Known Activations