INDEX

Explanations

negative evaluations or criticisms

oai_token-act-pair · gpt-3.5-turbo

the concept of criticism directed at various subjects

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

 reperto

-0.65

 istg

-0.62

igmatic

-0.61

 jewels

-0.61

 gasp

-0.60

ãĤ¼ãĤ¦ãĤ¹

-0.59

mAh

-0.59

fry

-0.59

 saliva

-0.58

 ILCS

-0.57

POSITIVE LOGITS

enance

0.81

 sorts

0.79

 course

0.79

 dissent

0.74

social

0.72

obin

0.72

 sentiment

0.71

speech

0.69

odox

0.69

science

0.69

Activations Density 0.122%

negative evaluations or criticisms

the concept of criticism directed at various subjects

No Comments

No Known Activations