INDEX

Explanations

words related to correctness, justification, or validation

oai_token-act-pair · gpt-3.5-turbo

words and phrases that convey a sense of correctness or justified actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

iments

-0.71

isms

-0.68

 Coffee

-0.68

 Alam

-0.67

 Football

-0.65

 shirts

-0.64

FM

-0.64

 fertility

-0.63

akeru

-0.63

 dolls

-0.63

POSITIVE LOGITS

ãĤ©

1.06

 rightly

0.99

 rightfully

0.97

 deserved

0.91

é¾į

0.87

è¯

0.82

 deserves

0.74

 outweigh

0.73

 deserve

0.72

æĺ¯

0.72

Activations Density 0.011%

words related to correctness, justification, or validation

words and phrases that convey a sense of correctness or justified actions

No Comments

No Known Activations