INDEX

Explanations

names of individuals in legal contexts

oai_token-act-pair · gpt-3.5-turbo

punctuation, specifically commas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 stereotypes

-0.69

FIX

-0.69

 tremend

-0.69

 dreams

-0.67

 predictions

-0.65

 incon

-0.64

Availability

-0.64

 prompt

-0.64

 animals

-0.63

 deficit

-0.63

POSITIVE LOGITS

Jr

1.18

Sr

1.09

tein

1.01

Jr

0.99

aka

0.95

LLP

0.92

QC

0.80

 Calif

0.78

Es

0.76

 supra

0.75

Activations Density 0.678%

names of individuals in legal contexts

punctuation, specifically commas

No Comments

No Known Activations