INDEX

Explanations

actions related to decision-making

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "choose to" indicating decision-making

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

til

-0.77

 Canary

-0.69

listed

-0.68

ILY

-0.68

 reperto

-0.64

inguished

-0.62

ritical

-0.62

pins

-0.62

 manuscripts

-0.61

 Five

-0.61

POSITIVE LOGITS

 pursue

1.08

 ignore

0.99

 make

0.99

 explore

0.97

 give

0.97

 maximize

0.96

 prioritize

0.96

 minimize

0.96

 create

0.94

 marry

0.94

Activations Density 0.067%

actions related to decision-making

instances of the phrase "choose to" indicating decision-making

No Comments

No Known Activations