INDEX

Explanations

keywords related to business, industries, and organizations

oai_token-act-pair · gpt-3.5-turbo

conjunctions and their contexts within sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

>[

-0.82

 oranges

-0.71

cells

-0.71

 Countries

-0.65

 bras

-0.64

 Forsaken

-0.64

edIn

-0.62

 disbel

-0.61

]}

-0.61

appropriately

-0.61

POSITIVE LOGITS

NG

0.74

NESS

0.72

election

0.70

visory

0.69

ifice

0.68

 preservation

0.67

rogen

0.66

marine

0.66

acle

0.65

wake

0.63

Activations Density 0.310%

keywords related to business, industries, and organizations

conjunctions and their contexts within sentences

No Comments

No Known Activations