INDEX

Explanations

instances of the word "reinforce" and its variations, indicating a focus on the concept of reinforcement

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

chens

-0.78

bang

-0.77

kie

-0.77

kers

-0.73

elle

-0.73

ascus

-0.72

few

-0.69

nia

-0.67

cas

-0.66

cture

-0.66

POSITIVE LOGITS

 reinforcing

1.09

 reinforcement

0.92

 reinforced

0.80

atively

0.71

ively

0.71

 falsehood

0.69

 layers

0.69

 suspicions

0.69

ments

0.68

 myths

0.67

Activations Density 0.016%

instances of the word "reinforce" and its variations, indicating a focus on the concept of reinforcement

No Comments

No Known Activations