INDEX

Explanations

phrases related to conflicts or confrontations

oai_token-act-pair · gpt-3.5-turbo

concepts related to intervention, conflict, and societal issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ĪĴ

-0.67

¡

-0.64

ĨĴ

-0.60

Ħ¢

-0.60

 princ

-0.56

Tes

-0.55

 teens

-0.53

ª

-0.53

 Kemp

-0.53

°

-0.52

POSITIVE LOGITS

0.98

.:

0.93

.''.

0.85

.;

0.83

.(

0.82

.#

0.76

.,

0.74

.-

0.72

.–

0.72

.?

0.71

Activations Density 0.255%

phrases related to conflicts or confrontations

concepts related to intervention, conflict, and societal issues

No Comments

No Known Activations