INDEX

Explanations

terms related to evaluating performance

oai_token-act-pair · gpt-3.5-turbo

phrases related to performance or effectiveness

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

:\

-0.76

ubs

-0.76

ãĤ´ãĥ³

-0.75

ãĥĩãĤ£

-0.74

ATOR

-0.73

Sov

-0.71

pez

-0.70

ERO

-0.69

eni

-0.69

iewicz

-0.69

POSITIVE LOGITS

 invested

0.73

 they

0.70

 compiled

0.70

 fired

0.70

idious

0.70

 pursu

0.67

 militar

0.67

 propelled

0.67

 elapsed

0.67

 awaited

0.66

Activations Density 0.039%

terms related to evaluating performance

phrases related to performance or effectiveness

No Comments

No Known Activations