INDEX

Explanations

mentions of corporate executives

oai_token-act-pair · gpt-3.5-turbo

references to corporate executives and their roles or actions

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

INGTON

-0.82

ttp

-0.72

¬¼

-0.70

DonaldTrump

-0.69

apo

-0.69

Temperature

-0.68

ãĥ¼ãĥ³

-0.68

ruck

-0.67

Operation

-0.66

Interstitial

-0.64

POSITIVE LOGITS

 executives

0.86

otonin

0.78

doms

0.77

who

0.72

 Doodle

0.69

obin

0.68

cius

0.68

 reap

0.67

 suites

0.67

paces

0.67

Activations Density 0.052%

mentions of corporate executives

references to corporate executives and their roles or actions

No Comments

No Known Activations