INDEX

Explanations

phrases related to negative events or experiences

oai_token-act-pair · gpt-3.5-turbo

references to criminal incidents and the implications surrounding them

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Reviewer

-0.99

storage

-0.63

 powering

-0.63

tons

-0.62

 directions

-0.62

oves

-0.62

Sold

-0.61

 Options

-0.61

Choice

-0.60

BAT

-0.60

POSITIVE LOGITS

 occur

1.90

 occurred

1.89

 happened

1.82

 happen

1.82

 happ

1.65

 occurs

1.52

 transpired

1.52

 Happ

1.45

 happens

1.42

 occurring

1.36

Activations Density 0.351%

phrases related to negative events or experiences

references to criminal incidents and the implications surrounding them

No Comments

No Known Activations