INDEX

Explanations

mentions of political figures and processes, especially related to elections and campaigns

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>

-3.12

 intersper

-1.37

/***

-1.28

 encomp

-1.10

 quitted

-1.09

 gratify

-1.03

 endow

-1.03

 vainly

-1.01

 effectually

-0.98

 disagre

-0.97

POSITIVE LOGITS

 Demok

0.92

 Muhamma

0.90

 silikon

0.89

 kosme

0.89

 mikrofon

0.89

uhr

0.86

 kado

0.86

 keramik

0.85

 optik

0.84

 Demokrat

0.83

Activations Density 1.817%

mentions of political figures and processes, especially related to elections and campaigns

No Comments

No Known Activations