INDEX

Explanations

instances where claims made are unsupported or unconfirmed, especially when related to facts or sources

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

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tricot

-1.02

 fluo

-0.80

 teras

-0.70

 oleo

-0.68

 centrif

-0.67

 fers

-0.66

 matel

-0.66

 pommes

-0.66

 Romains

-0.66

 parf

-0.66

POSITIVE LOGITS

 truth

0.71

 verified

0.71

 veracity

0.68

 verify

0.68

 verification

0.67

 verifiable

0.67

 factual

0.67

 credible

0.64

 verifying

0.64

 credibility

0.63

Activations Density 0.710%

instances where claims made are unsupported or unconfirmed, especially when related to facts or sources

No Comments

No Known Activations