INDEX

Explanations

specific instances of the word "hired"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "hired."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

edly

-0.74

Bah

-0.67

 Strait

-0.66

otropic

-0.65

ahi

-0.65

uclear

-0.64

alone

-0.63

ican

-0.61

EVA

-0.61

eq

-0.60

POSITIVE LOGITS

 hired

0.90

lings

0.87

 assistants

0.82

glers

0.82

æ©

0.82

 hire

0.75

ãĤŃ

0.74

 interns

0.73

 Dame

0.70

 mercenaries

0.69

Activations Density 0.014%

specific instances of the word "hired"

instances of the word "hired."

No Comments

No Known Activations