INDEX

Explanations

mentions of the name "io" at varying activation levels

oai_token-act-pair · gpt-3.5-turbo

references to the programming language "Io."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

holder

-0.79

ancies

-0.73

ivities

-0.69

rap

-0.66

puter

-0.66

rier

-0.65

uates

-0.64

atche

-0.63

 ILCS

-0.62

 glim

-0.60

POSITIVE LOGITS

ppo

1.17

cean

1.06

zzi

1.03

ctl

1.02

ption

0.97

ffic

0.93

active

0.93

vernment

0.91

hazard

0.90

pec

0.87

Activations Density 0.027%

mentions of the name "io" at varying activation levels

references to the programming language "Io."

No Comments

No Known Activations