INDEX

Explanations

terms related to peaks or extremes

oai_token-act-pair · gpt-3.5-turbo

references to extremes in prices or performance metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

OTS

-0.75

Guide

-0.73

 Galile

-0.69

Austral

-0.69

Fac

-0.68

shapeshifter

-0.68

 Caucas

-0.68

RECT

-0.68

Reply

-0.66

Shell

-0.66

POSITIVE LOGITS

 highs

1.17

 lows

1.05

terday

0.98

terness

0.96

mith

0.90

 heights

0.87

imes

0.80

creen

0.79

daq

0.79

imar

0.78

Activations Density 0.017%

terms related to peaks or extremes

references to extremes in prices or performance metrics

No Comments

No Known Activations