INDEX

Explanations

code snippets and technical instructions

oai_token-act-pair · gpt-3.5-turbo

general performance indicators or metrics across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 diving

-0.73

 occas

-0.72

 undermin

-0.71

 scrambling

-0.71

 rooting

-0.70

 dodging

-0.69

 freel

-0.69

 hopping

-0.68

 rallying

-0.67

 winding

-0.65

POSITIVE LOGITS

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

1.12

âĵĺ

1.08

SELECT

1.08

Products

1.07

ÂłÂłÂłÂł

1.05

ÂłÂłÂł

1.04

RESULTS

1.04

posted

1.02

Overview

1.00

Example

1.00

Activations Density 0.407%

code snippets and technical instructions

general performance indicators or metrics across various contexts

No Comments

No Known Activations