INDEX

Explanations

references to the Internal Revenue Service (IRS)

oai_token-act-pair · gpt-3.5-turbo

mentions of the IRS (Internal Revenue Service) and associated contexts

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

boards

-0.85

axter

-0.77

izations

-0.77

hots

-0.74

theless

-0.73

ifest

-0.71

isations

-0.71

McA

-0.66

 Kard

-0.66

 Jenner

-0.65

POSITIVE LOGITS

IRS

0.96

 scrutiny

0.77

 levied

0.74

 Oversight

0.73

 Revenue

0.72

 scandal

0.72

 Lerner

0.72

eteria

0.72

 stalking

0.71

 bureaucracy

0.71

Activations Density 0.019%

references to the Internal Revenue Service (IRS)

mentions of the IRS (Internal Revenue Service) and associated contexts

No Comments

No Known Activations