INDEX

Explanations

words related to negative judgments or evaluations

oai_token-act-pair · gpt-3.5-turbo

expressions of negative judgments or sentiments, particularly the word "terrible."

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

ership

-0.86

pai

-0.86

irs

-0.85

ilus

-0.80

aver

-0.75

eters

-0.73

cript

-0.72

paio

-0.72

izen

-0.71

gat

-0.70

POSITIVE LOGITS

 sounding

0.81

 havoc

0.78

NESS

0.77

 awful

0.77

 nightmares

0.76

 horrible

0.75

 headache

0.74

 nightmare

0.73

 adolesc

0.72

 ordeal

0.72

Activations Density 0.022%

words related to negative judgments or evaluations

expressions of negative judgments or sentiments, particularly the word "terrible."

No Comments

No Known Activations