INDEX

Explanations

phrases related to interaction or connection with external entities, such as reaching out, getting involved, or seeking feedback

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>

-2.26

 intersper

-0.77

 endow

-0.76

 strove

-0.76

 endeavoured

-0.76

 strode

-0.75

 winked

-0.75

 endeavored

-0.73

 Lmao

-0.73

 disbur

-0.73

POSITIVE LOGITS

 véhic

0.83

ereka

0.83

ouret

0.82

 proprement

0.79

 dégust

0.78

siyah

0.76

 goû

0.75

ughter

0.74

 catég

0.73

 préc

0.73

Activations Density 0.378%

phrases related to interaction or connection with external entities, such as reaching out, getting involved, or seeking feedback

No Comments

No Known Activations