INDEX

Explanations

phrases related to current events and politics, particularly focusing on controversial topics

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

weights

-0.74

 decomp

-0.73

conservancy

-0.72

 bung

-0.69

 hosting

-0.69

 clad

-0.68

 nude

-0.68

 descended

-0.67

 electing

-0.67

 envelop

-0.67

POSITIVE LOGITS

ONSORED

1.20

 Unfortunately

0.98

 Additionally

0.93

However

0.93

 Conversely

0.92

 Also

0.92

 Similarly

0.92

 However

0.91

 Likewise

0.89

 Besides

0.88

Activations Density 1.185%

phrases related to current events and politics, particularly focusing on controversial topics

No Comments

No Known Activations