INDEX

Explanations

occurrences of the word "loss" with a focus on negative consequences

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the term "loss."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ENTS

-0.70

rouse

-0.66

enegger

-0.66

ATT

-0.65

ECK

-0.64

 populated

-0.63

 Caucas

-0.62

ansky

-0.62

chell

-0.59

POSITIVE LOGITS

 loss

1.05

 aversion

1.05

 Loss

0.98

loss

0.93

 losses

0.88

iem

0.87

luster

0.75

 prevention

0.73

lust

0.73

front

0.71

Activations Density 0.013%

occurrences of the word "loss" with a focus on negative consequences

repeated mentions of the term "loss."

No Comments

No Known Activations