INDEX

Explanations

phrases related to suspicion or suspicion-based actions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "suspicion" and related terms indicating doubt or distrust

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hire

-0.73

char

-0.73

options

-0.73

dos

-0.72

jri

-0.70

umbn

-0.69

itters

-0.69

church

-0.67

yss

-0.67

alloc

-0.67

POSITIVE LOGITS

lessly

1.03

 suspicion

0.99

 suspicions

0.83

less

0.79

lessness

0.74

 suspects

0.72

ãĥĨ

0.70

 suspicious

0.68

 warrant

0.68

worthiness

0.65

Activations Density 0.017%

phrases related to suspicion or suspicion-based actions

instances of the word "suspicion" and related terms indicating doubt or distrust

No Comments

No Known Activations