INDEX

Explanations

terms related to skills, expertise, ideas, and behaviors

oai_token-act-pair · gpt-3.5-turbo

terms related to expertise and professional skills

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

$.

-0.70

omever

-0.69

 oneself

-0.68

!.

-0.64

QUIRE

-0.64

orks

-0.64

È

-0.63

endez

-0.61

Common

-0.61

 Pearce

-0.59

POSITIVE LOGITS

 consisted

0.93

 consists

0.88

 woes

0.82

 differs

0.74

 revolves

0.73

 encompasses

0.72

 companion

0.72

 rede

0.71

 boils

0.71

 extends

0.70

Activations Density 0.494%

terms related to skills, expertise, ideas, and behaviors

terms related to expertise and professional skills

No Comments

No Known Activations