INDEX

Explanations

organizations or entities that are identified as "think tanks."

oai_token-act-pair · gpt-3.5-turbo

terms related to political ideologies and think tanks

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

Ô

-0.90

theless

-0.82

 Helpful

-0.81

 Detected

-0.77

Output

-0.74

TPPStreamerBot

-0.69

 Towards

-0.67

[+]

-0.65

 Decay

-0.64

 attribution

-0.64

POSITIVE LOGITS

ocrin

0.83

arist

0.76

 camp

0.76

ops

0.75

agent

0.75

 estate

0.73

mans

0.72

fund

0.72

inder

0.71

intern

0.71

Activations Density 0.498%

organizations or entities that are identified as "think tanks."

terms related to political ideologies and think tanks

No Comments

No Known Activations