INDEX

Explanations

instances of destruction, closure, or negative impact related to various situations or establishments

oai_token-act-pair · gpt-3.5-turbo

words associated with destruction and loss

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

)</

-0.61

osite

-0.58

osit

-0.57

ocent

-0.57

FK

-0.57

 Badge

-0.55

{:

-0.54

DCS

-0.54

 answ

-0.53

 resemblance

-0.53

POSITIVE LOGITS

 prematurely

0.91

 amid

0.86

 overnight

0.84

 unnecessarily

0.83

or

0.82

 altogether

0.79

 wholesale

0.77

 indiscrim

0.76

by

0.75

 abruptly

0.74

Activations Density 0.240%

instances of destruction, closure, or negative impact related to various situations or establishments

words associated with destruction and loss

No Comments

No Known Activations