INDEX

Explanations

numbers, measurements, and quantifiable data

oai_token-act-pair · gpt-3.5-turbo

numerical data and references to statistical measures or evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

swick

-0.75

kees

-0.73

friends

-0.72

rities

-0.71

Others

-0.67

owners

-0.67

;;;;;;;;

-0.65

ggies

-0.64

deals

-0.64

raid

-0.64

POSITIVE LOGITS

 predictive

0.88

 coefficient

0.85

 predicting

0.81

 Celsius

0.80

 predictor

0.80

 probability

0.75

 acronym

0.75

 decimal

0.75

 algorithm

0.74

 coefficients

0.73

Activations Density 1.148%

numbers, measurements, and quantifiable data

numerical data and references to statistical measures or evidence

No Comments

No Known Activations