INDEX

Explanations

references to accidents, incidents involving harm, and violent events

oai_token-act-pair · gpt-3.5-turbo

occurrences of violent incidents or deaths

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Pros

-0.72

hett

-0.71

}}}

-0.70

plom

-0.67

eworld

-0.64

ecd

-0.64

Stud

-0.63

SEC

-0.61

Intern

-0.61

apps

-0.61

POSITIVE LOGITS

 spree

1.08

 involving

0.99

 occurred

0.90

 perpetrated

0.87

 altercation

0.84

 rampage

0.83

 burglary

0.79

 erupted

0.76

 unrelated

0.76

 explosion

0.76

Activations Density 0.258%

references to accidents, incidents involving harm, and violent events

occurrences of violent incidents or deaths

No Comments

No Known Activations