INDEX

Explanations

mentions of stock market symbols or financial institutions

oai_token-act-pair · gpt-3.5-turbo

references to stock market indexes and ticker symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

packs

-0.68

wic

-0.64

hur

-0.62

packing

-0.61

 drunken

-0.61

pack

-0.60

ophile

-0.59

 kits

-0.58

ordan

-0.58

cru

-0.58

POSITIVE LOGITS

DAQ

1.74

daq

1.73

NYSE

1.02

ogle

0.81

enta

0.79

ãĤ´ãĥ³

0.76

=#

0.76

 Composite

0.76

 Trading

0.76

alore

0.73

Activations Density 0.004%

mentions of stock market symbols or financial institutions

references to stock market indexes and ticker symbols

No Comments

No Known Activations