INDEX

Explanations

accusations of wrongdoing or misconduct

oai_token-act-pair · gpt-3.5-turbo

instances of the word "accused" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

emetery

-0.89

ARCH

-0.75

 tremend

-0.71

ental

-0.69

ECT

-0.68

ONSORED

-0.68

atial

-0.67

erved

-0.67

arp

-0.66

hare

-0.66

POSITIVE LOGITS

 accuse

0.95

 accusing

0.89

 accuses

0.85

ologists

0.79

 abusers

0.78

isance

0.77

 accused

0.74

 betray

0.72

 accusations

0.71

ologies

0.71

Activations Density 0.033%

accusations of wrongdoing or misconduct

instances of the word "accused" in various contexts

No Comments

No Known Activations