INDEX

Explanations

words related to degradation or decline

oai_token-act-pair · gpt-3.5-turbo

terms related to degradation or decline

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

ãĥĹ

-0.75

ites

-0.72

 GOODMAN

-0.71

ohyd

-0.69

éĹĺ

-0.67

enegger

-0.67

ãĥĦ

-0.67

FTWARE

-0.67

 PsyNet

-0.65

tery

-0.64

POSITIVE LOGITS

gradation

1.21

rees

1.04

asus

0.98

mentation

0.98

ener

0.97

rowth

0.97

rad

0.92

deg

0.91

rade

0.89

 degradation

0.88

Activations Density 0.012%

words related to degradation or decline

terms related to degradation or decline

No Comments

No Known Activations