INDEX

Explanations

words related to providing clarification, legal assistance, and assistance in various forms

oai_token-act-pair · gpt-3.5-turbo

references to the concept of relief in various contexts

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

tein

-0.80

 Lethal

-0.75

0000000000000000

-0.73

eton

-0.72

 Wilde

-0.69

Phi

-0.69

prem

-0.63

 FANT

-0.63

 Bulg

-0.63

Ô

-0.63

POSITIVE LOGITS

 relief

1.23

 Relief

0.98

ilitation

0.94

 overboard

0.76

ilitating

0.76

 valve

0.75

 pitcher

0.74

 relieve

0.73

 pitchers

0.71

iants

0.71

Activations Density 0.008%

words related to providing clarification, legal assistance, and assistance in various forms

references to the concept of relief in various contexts

No Comments

No Known Activations