INDEX

Explanations

occurrences of the word "called"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bilt

-0.70

 isolate

-0.69

Hispanic

-0.69

inth

-0.69

 isolated

-0.67

olitics

-0.66

agall

-0.66

jen

-0.65

SPONSORED

-0.64

 drained

-0.64

POSITIVE LOGITS

"#

0.85

 Operation

0.84

 Nightmares

0.81

 Calling

0.79

 Sleeping

0.78

igraph

0.77

 Attention

0.73

 Lethal

0.72

 Plays

0.72

 Fatal

0.70

Activations Density 0.026%

occurrences of the word "called"

No Comments

No Known Activations