INDEX

Explanations

proper nouns that seem to be related to political or social contexts

oai_token-act-pair · gpt-3.5-turbo

prominent people, organizations, and geopolitical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Crescent

-0.50

OURCE

-0.49

 Curious

-0.48

ULL

-0.48

Comments

-0.48

 unden

-0.48

luster

-0.47

++++++++

-0.47

.#

-0.47

+.

-0.46

POSITIVE LOGITS

 didnt

0.85

 hadn

0.84

 forgot

0.81

 could

0.80

 doesnt

0.78

had

0.78

 cannot

0.76

 deserved

0.75

 should

0.75

 lacks

0.74

Activations Density 0.585%

proper nouns that seem to be related to political or social contexts

prominent people, organizations, and geopolitical references

No Comments

No Known Activations