INDEX

Explanations

words related to potions or magical elixirs

oai_token-act-pair · gpt-3.5-turbo

references to a specific term or concept related to intellectual elitism

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

ãĥ¼ãĥĨãĤ£

-0.83

NEY

-0.77

ãĥ´ãĤ¡

-0.76

 Crunch

-0.74

 Freeze

-0.68

 Analytics

-0.67

SHIP

-0.67

 Accountability

-0.66

apeake

-0.65

nant

-0.65

POSITIVE LOGITS

izabeth

1.17

iquid

1.15

ibrary

1.13

ixir

1.02

usive

0.98

uding

0.97

uded

0.95

abor

0.90

ittle

0.89

azar

0.88

Activations Density 0.010%

words related to potions or magical elixirs

references to a specific term or concept related to intellectual elitism

No Comments

No Known Activations