INDEX

Explanations

text related to leaked information, particularly in a media or news context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 disagre

-1.71

 increa

-1.64

 encomp

-1.61

 emphat

-1.55

 impra

-1.55

 reluct

-1.52

 depic

-1.50

 affor

-1.48

 milf

-1.48

 shenan

-1.44

POSITIVE LOGITS

 disclosure

0.86

 public

0.84

 leak

0.78

 disclose

0.77

public

0.77

pub

0.76

 disclosed

0.75

 publication

0.75

 publicly

0.74

 gepubliceerd

0.72

Activations Density 0.467%

text related to leaked information, particularly in a media or news context

No Comments

No Known Activations