INDEX

Explanations

words related to summaries or collections of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 Patriarch

-0.71

Mesh

-0.70

 Strongh

-0.67

 mortg

-0.65

 constructs

-0.63

 Strait

-0.62

ippi

-0.62

 Formation

-0.62

 Myster

-0.62

lvl

-0.60

POSITIVE LOGITS

 roundup

0.88

ãĥ¼ãĥ³

0.88

icles

0.79

ashington

0.78

weekly

0.77

icle

0.75

ãĥĥãĤ¯

0.74

govtrack

0.73

efully

0.69

ãĥ¢

0.69

Activations Density 0.005%

words related to summaries or collections of information

No Comments

No Known Activations