INDEX

Explanations

information related to physical attributes like height and weight

oai_token-act-pair · gpt-3.5-turbo

weight and height specifications

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

urtles

-0.68

 Solitaire

-0.64

riet

-0.60

rant

-0.59

 satisfaction

-0.58

uca

-0.57

 plur

-0.57

atis

-0.57

 Lilly

-0.57

 worsened

-0.57

POSITIVE LOGITS

0.90

0.89

0.87

0.86

0.85

0.82

Activations Density 0.079%

information related to physical attributes like height and weight

weight and height specifications

No Comments

No Known Activations