INDEX

Explanations

expressions of dialogue, particularly those indicating emphasis or strong feelings

oai_token-act-pair · gpt-3.5-turbo

speak about individuals or groups in a derogatory or condescending manner

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Seym

-0.75

 mathemat

-0.69

 tabloid

-0.68

 limb

-0.68

 ivory

-0.68

 seiz

-0.67

 Gardens

-0.67

 amusement

-0.66

 metic

-0.66

 tasting

-0.66

POSITIVE LOGITS

ï¸ı

1.06

rd

0.99

lean

0.95

resent

0.93

deg

0.89

vernment

0.89

ï¸

0.85

PB

0.84

audi

0.84

ãĥĥãĥī

0.84

Activations Density 0.030%

expressions of dialogue, particularly those indicating emphasis or strong feelings

speak about individuals or groups in a derogatory or condescending manner

No Comments

No Known Activations