INDEX

Explanations

words related to triggers or initiating events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "trigger" and its variations indicating causation or initiation of events

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

hemat

-0.80

ensable

-0.73

 Cutler

-0.68

 Expend

-0.68

esan

-0.67

ately

-0.67

iance

-0.64

aredevil

-0.64

iago

-0.64

ibble

-0.63

POSITIVE LOGITS

 triggering

1.13

trigger

1.12

 triggers

1.06

 trigger

1.05

 Trigger

0.88

 triggered

0.86

witz

0.83

 warnings

0.82

chnology

0.79

Trigger

0.78

Activations Density 0.018%

words related to triggers or initiating events

occurrences of the word "trigger" and its variations indicating causation or initiation of events

No Comments

No Known Activations