INDEX

Explanations

information related to variability and differences in various contexts

oai_token-act-pair · gpt-3.5-turbo

concepts related to variability and differences among categories

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

DEF

-0.72

ãĤ¸

-0.71

iron

-0.69

ãĥį

-0.66

oho

-0.64

 Damon

-0.64

Inside

-0.63

Alert

-0.63

WARN

-0.61

Lost

-0.61

POSITIVE LOGITS

 differing

1.26

different

1.10

 sizes

1.09

 different

1.09

 climates

1.05

Different

1.04

 varying

1.03

 Different

1.00

 genders

0.99

 severity

0.98

Activations Density 0.543%

information related to variability and differences in various contexts

concepts related to variability and differences among categories

No Comments

No Known Activations