INDEX

Explanations

correlations between different variables or factors

oai_token-act-pair · gpt-3.5-turbo

relationships between variables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

lite

-0.69

////////////////////////////////

-0.66

sung

-0.65

Flo

-0.65

OWN

-0.64

zed

-0.62

AND

-0.60

skirts

-0.59

 Surv

-0.58

mbuds

-0.56

POSITIVE LOGITS

 between

1.34

between

1.24

 Between

1.17

 linking

1.01

uality

1.01

SHIP

1.01

BET

0.97

alogy

0.87

ality

0.83

ability

0.82

Activations Density 0.073%

correlations between different variables or factors

relationships between variables

No Comments

No Known Activations