INDEX

Explanations

mentions of physical harm or injury

oai_token-act-pair · gpt-3.5-turbo

references to injuries and their consequences

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

soDeliveryDate

-1.07

sonian

-0.93

akeru

-0.86

eu

-0.82

African

-0.78

 acad

-0.78

ħĭ

-0.77

urated

-0.76

elligent

-0.75

wcsstore

-0.75

POSITIVE LOGITS

 injuries

0.96

 injury

0.94

 dehydration

0.90

 incurred

0.90

 inhal

0.86

 untreated

0.84

 illness

0.83

 inflicted

0.82

 ailments

0.81

 coma

0.80

Activations Density 0.564%

mentions of physical harm or injury

references to injuries and their consequences

No Comments

No Known Activations