INDEX

Explanations

phrases related to criticism or discrediting

oai_token-act-pair · gpt-3.5-turbo

instances of accountability and judgment in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ´ãĥ³

-1.01

ãĤ¨ãĥ«

-0.85

ãģł

-0.80

ãĤ¦ãĤ¹

-0.80

rossover

-0.77

visory

-0.75

igun

-0.74

iHUD

-0.73

EngineDebug

-0.68

ãĥ£

-0.67

POSITIVE LOGITS

 policymakers

0.97

 researchers

0.96

we

0.96

 however

0.95

 designers

0.91

 they

0.91

 proponents

0.86

 filmmakers

0.86

 attackers

0.86

 readers

0.85

Activations Density 0.231%

phrases related to criticism or discrediting

instances of accountability and judgment in various contexts

No Comments

No Known Activations