INDEX

Explanations

references to attacks or threats of violence

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

 للاسماء

-0.52

JspWriter

-0.48

setPointSize

-0.46

 prome

-0.46

GTCX

-0.46

UniformLocation

-0.44

Décès

-0.43

FunctionFlags

-0.42

 Grund

-0.42

 beig

-0.42

POSITIVE LOGITS

 attack

1.25

 Attack

1.16

attack

1.14

 attacks

1.10

 Attacks

1.08

Attack

1.04

Attacks

1.00

attacks

0.99

 attacked

0.98

 attacking

0.97

Activations Density 0.107%

references to attacks or threats of violence

No Comments

No Known Activations